2020.03.25.txt

==========New Papers==========
1, TITLE:       Real-Time Dispatching of Large-Scale Ride-Sharing Systems: Integrating Optimization, Machine Learning, and Model Predictive Control
http://arxiv.org/abs/2003.10942
AUTHORS:        Connor Riley ; Pascal Van Hentenryck ; Enpeng Yuan
HIGHLIGHT:      This paper proposes an end-to-end approach that tightly integrates a state-of-the-art dispatching algorithm, a machine-learning model to predict zone-to-zone demand over time, and a model predictive control optimization to relocate idle vehicles.

2, TITLE:       Learning Compact Reward for Image Captioning
http://arxiv.org/abs/2003.10925
AUTHORS:        Nannan Li ; Zhenzhong Chen
COMMENTS:       13 pages, 10 figures
HIGHLIGHT:      In this paper, we propose a refined Adversarial Inverse Reinforcement Learning (rAIRL) method to handle the reward ambiguity problem by disentangling reward for each word in a sentence, as well as achieve stable adversarial training by refining the loss function to shift the generator towards Nash equilibrium.

3, TITLE:       RN-VID: A Feature Fusion Architecture for Video Object Detection
http://arxiv.org/abs/2003.10898
AUTHORS:        Hughes Perreault ; Maguelonne Héritier ; Pierre Gravel ; Guillaume-Alexandre Bilodeau ; Nicolas Saunier
HIGHLIGHT:      It is with this idea in mind that we propose RN-VID, a novel approach to video object detection.

4, TITLE:       Registration by tracking for sequential 2D MRI
http://arxiv.org/abs/2003.10819
AUTHORS:        Niklas Gunnarsson ; Jens Sjölund ; Thomas B. Schön
COMMENTS:       Currently under review for a conference
HIGHLIGHT:      In this paper we present an image registration method that exploits the sequential nature of 2D MR images to estimate the corresponding displacement field.

5, TITLE:       Multi-Scale Progressive Fusion Network for Single Image Deraining
http://arxiv.org/abs/2003.10985
AUTHORS:        Kui Jiang ; Zhongyuan Wang ; Peng Yi ; Chen Chen ; Baojin Huang ; Yimin Luo ; Jiayi Ma ; Junjun Jiang
COMMENTS:       CVPR 2020
HIGHLIGHT:      In this work, we explore the multi-scale collaborative representation for rain streaks from the perspective of input image scales and hierarchical deep features in a unified framework, termed multi-scale progressive fusion network (MSPFN) for single image rain streak removal.

6, TITLE:       Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstruction
http://arxiv.org/abs/2003.10983
AUTHORS:        Rohan Chabra ; Jan Eric Lenssen ; Eddy Ilg ; Tanner Schmidt ; Julian Straub ; Steven Lovegrove ; Richard Newcombe
HIGHLIGHT:      To address this problem we introduce Deep Local Shapes (DeepLS), a deep shape representation that enables encoding and reconstruction of high-quality 3D shapes without prohibitive memory requirements.

7, TITLE:       Generalizing Spatial Transformers to Projective Geometry with Applications to 2D/3D Registration
http://arxiv.org/abs/2003.10987
AUTHORS:        Cong Gao ; Xingtong Liu ; Wenhao Gu ; Benjamin Killeen ; Mehran Armand ; Russell Taylor ; Mathias Unberath
HIGHLIGHT:      We propose a novel Projective Spatial Transformer module that generalizes spatial transformers to projective geometry, thus enabling differentiable volume rendering.

8, TITLE:       Know Your Surroundings: Exploiting Scene Information for Object Tracking
http://arxiv.org/abs/2003.11014
AUTHORS:        Goutam Bhat ; Martin Danelljan ; Luc Van Gool ; Radu Timofte
HIGHLIGHT:      In this work, we propose a novel tracking architecture which can utilize scene information for tracking.

9, TITLE:       Exploiting Event Cameras by Using a Network Grafting Algorithm
http://arxiv.org/abs/2003.10959
AUTHORS:        Yuhuang Hu ; Tobi Delbruck ; Shih-Chii Liu
HIGHLIGHT:      This paper proposes a Network Grafting Algorithm (NGA), where a new front end network driven by unconventional visual inputs replaces the front end network of a pretrained deep network that processes intensity frames.

10, TITLE:      MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Mask
http://arxiv.org/abs/2003.10955
AUTHORS:        Shengyu Zhao ; Yilun Sheng ; Yue Dong ; Eric I-Chao Chang ; Yan Xu
COMMENTS:       CVPR 2020 (Oral)
HIGHLIGHT:      In this paper, we propose an asymmetric occlusion-aware feature matching module, which can learn a rough occlusion mask that filters useless (occluded) areas immediately after feature warping without any explicit supervision.

11, TITLE:      Investigating Software Usage in the Social Sciences: A Knowledge Graph Approach
http://arxiv.org/abs/2003.10715
AUTHORS:        David Schindler ; Benjamin Zapilko ; Frank Krüger
COMMENTS:       16 pages, 4 figures, preprint of a full paper at Extended Semantic Web Conference (ESWC 2020)
HIGHLIGHT:      In this paper, we present SoftwareKG - a knowledge graph that contains information about software mentions from more than 51,000 scientific articles from the social sciences.

12, TITLE:      Generating Chinese Poetry from Images via Concrete and Abstract Information
http://arxiv.org/abs/2003.10773
AUTHORS:        Yusen Liu ; Dayiheng Liu ; Jiancheng Lv ; Yongsheng Sang
COMMENTS:       Accepted by the 2020 International Joint Conference on Neural Networks (IJCNN 2020)
HIGHLIGHT:      In this paper, we extract and integrate the Concrete and Abstract information from images to address those issues.

13, TITLE:      Felix: Flexible Text Editing Through Tagging and Insertion
http://arxiv.org/abs/2003.10687
AUTHORS:        Jonathan Mallinson ; Aliaksei Severyn ; Eric Malmi ; Guillermo Garrido
HIGHLIGHT:      We present Felix --- a flexible text-editing approach for generation, designed to derive the maximum benefit from the ideas of decoding with bi-directional contexts and self-supervised pre-training.

14, TITLE:      Towards Neural Machine Translation for Edoid Languages
http://arxiv.org/abs/2003.10704
AUTHORS:        Iroro Orife
COMMENTS:       Accepted to ICLR 2020 AfricaNLP workshop
HIGHLIGHT:      Using the new JW300 public dataset, we trained and evaluated baseline translation models for four widely spoken languages in this group: \`Ed\'o, \'Es\'an, Urhobo and Isoko.

15, TITLE:      Cross-Lingual Adaptation Using Universal Dependencies
http://arxiv.org/abs/2003.10816
AUTHORS:        Nasrin Taghizadeh ; Heshaam Faili
HIGHLIGHT:      We describe a cross-lingual adaptation method based on syntactic parse trees obtained from the Universal Dependencies (UD), which are consistent across languages, to develop classifiers in low-resource languages.

16, TITLE:      Modeling Contrary-to-Duty with CP-nets
http://arxiv.org/abs/2003.10480
AUTHORS:        Roberta Calegari ; Andrea Loreggia ; Emiliano Lorini ; Francesca Rossi ; Giovanni Sartor
HIGHLIGHT:      This paper shows how deontic concepts can be captured through conditional preference models.

17, TITLE:      ScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation
http://arxiv.org/abs/2003.10557
AUTHORS:        Sharon Fogel ; Hadar Averbuch-Elor ; Sarel Cohen ; Shai Mazor ; Roee Litman
COMMENTS:       in CVPR 2020
HIGHLIGHT:      We present ScrabbleGAN, a semi-supervised approach to synthesize handwritten text images that are versatile both in style and lexicon.

18, TITLE:      Label Noise Types and Their Effects on Deep Learning
http://arxiv.org/abs/2003.10471
AUTHORS:        Görkem Algan ; İlkay Ulusoy
HIGHLIGHT:      In this work, we provide a detailed analysis of the effects of different kinds of label noise on learning.

19, TITLE:      Broad Area Search and Detection of Surface-to-Air Missile Sites Using Spatial Fusion of Component Object Detections from Deep Neural Networks
http://arxiv.org/abs/2003.10566
AUTHORS:        Alan B. Cannaday II ; Curt H. Davis ; Grant J. Scott ; Blake Ruprecht ; Derek T. Anderson
COMMENTS:       9 pages, 9 figures, 9 tables, pre-published expansion of IGARSS2019 conference paper "Improved Search and Detection of Surface-to-Air Missile Sites Using Spatial Fusion of Component Object Detections from Deep Neural Networks"
HIGHLIGHT:      Here we demonstrate how Deep Neural Network (DNN) detections of multiple constitutive or component objects that are part of a larger, more complex, and encompassing feature can be spatially fused to improve the search, detection, and retrieval (ranking) of the larger complex feature.

20, TITLE:      Peeking into occluded joints: A novel framework for crowd pose estimation
http://arxiv.org/abs/2003.10506
AUTHORS:        Lingteng Qiu ; Xuanye Zhang ; Yanran Li ; Guanbin Li ; Xiaojun Wu ; Zixiang Xiong ; Xiaoguang Han ; Shuguang Cui
COMMENTS:       The code of OPEC-Net is available at: https://lingtengqiu.github.io/2020/03/22/OPEC-Net/
HIGHLIGHT:      Therefore, we thoroughly pursue this problem and propose a novel OPEC-Net framework together with a new Occluded Pose (OCPose) dataset with 9k annotated images.

21, TITLE:      Distillating Knowledge from Graph Convolutional Networks
http://arxiv.org/abs/2003.10477
AUTHORS:        Yiding Yang ; Jiayan Qiu ; Mingli Song ; Dacheng Tao ; Xinchao Wang
HIGHLIGHT:      In this paper, we propose to our best knowledge the first dedicated approach to {distilling} knowledge from a pre-trained GCN model.

22, TITLE:      ProGraML: Graph-based Deep Learning for Program Optimization and Analysis
http://arxiv.org/abs/2003.10536
AUTHORS:        Chris Cummins ; Zacharias V. Fisches ; Tal Ben-Nun ; Torsten Hoefler ; Hugh Leather
COMMENTS:       20 pages, author preprint
HIGHLIGHT:      We introduce ProGraML - Program Graphs for Machine Learning - a novel graph-based program representation using a low level, language agnostic, and portable format; and machine learning models capable of performing complex downstream tasks over these graphs.

23, TITLE:      Data-driven models and computational tools for neurolinguistics: a language technology perspective
http://arxiv.org/abs/2003.10540
AUTHORS:        Ekaterina Artemova ; Amir Bakarov ; Aleksey Artemov ; Evgeny Burnaev ; Maxim Sharaev
COMMENTS:       37 pages, 1 figure
HIGHLIGHT:      In this paper, our focus is the connection and influence of language technologies on the research in neurolinguistics.

24, TITLE:      Neural Networks and Polynomial Regression. Demystifying the Overparametrization Phenomena
http://arxiv.org/abs/2003.10523
AUTHORS:        Matt Emschwiller ; David Gamarnik ; Eren C. Kızıldağ ; Ilias Zadik
COMMENTS:       59 pages, 3 figures
HIGHLIGHT:      In this paper we prove a series of results which provide a somewhat diverging explanation.

25, TITLE:      Synergic Adversarial Label Learning with DR and AMD for Retinal Image Grading
http://arxiv.org/abs/2003.10607
AUTHORS:        Lie Ju ; Xin Wang ; Paul Bonnington ; Zongyuan Ge
HIGHLIGHT:      We propose a method called synergic adversarial label learning (SALL) which leverages relevant retinal disease labels in both semantic and feature space as additional signals and train the model in a collaborative manner.

26, TITLE:      Spatio-Temporal Handwriting Imitation
http://arxiv.org/abs/2003.10593
AUTHORS:        Martin Mayr ; Martin Stumpf ; Anguelos Nikolaou ; Mathias Seuret ; Andreas Maier ; Vincent Christlein
COMMENTS:       Main paper: 14 pages, supplemental material: 8 pages
HIGHLIGHT:      We show that subdividing the process into smaller subtasks makes it possible to imitate someone's handwriting with a high chance to be visually indistinguishable for humans.

27, TITLE:      KFNet: Learning Temporal Camera Relocalization using Kalman Filtering
http://arxiv.org/abs/2003.10629
AUTHORS:        Lei Zhou ; Zixin Luo ; Tianwei Shen ; Jiahui Zhang ; Mingmin Zhen ; Yao Yao ; Tian Fang ; Long Quan
COMMENTS:       An oral paper of CVPR 2020
HIGHLIGHT:      In this work, we improve the temporal relocalization method by using a network architecture that incorporates Kalman filtering (KFNet) for online camera relocalization.

28, TITLE:      UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World
http://arxiv.org/abs/2003.10608
AUTHORS:        Shangbang Long ; Cong Yao
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In this paper, we introduce UnrealText, an efficient image synthesis method that renders realistic images via a 3D graphics engine.

29, TITLE:      Video Object Grounding using Semantic Roles in Language Description
http://arxiv.org/abs/2003.10606
AUTHORS:        Arka Sadhu ; Kan Chen ; Ram Nevatia
COMMENTS:       CVPR20 camera-ready including appendix
HIGHLIGHT:      Here, we investigate the role of object relations in VOG and propose a novel framework VOGNet to encode multi-modal object relations via self-attention with relative position encoding. To evaluate VOGNet, we propose novel contrasting sampling methods to generate more challenging grounding input samples, and construct a new dataset called ActivityNet-SRL (ASRL) based on existing caption and grounding datasets.

30, TITLE:      Adversarial Perturbations Fool Deepfake Detectors
http://arxiv.org/abs/2003.10596
AUTHORS:        Apurva Gandhi ; Shomik Jain
COMMENTS:       To appear in the proceedings of the International Joint Conference on Neural Networks (IJCNN 2020)
HIGHLIGHT:      This work uses adversarial perturbations to enhance deepfake images and fool common deepfake detectors.

31, TITLE:      First Investigation Into the Use of Deep Learning for Continuous Assessment of Neonatal Postoperative Pain
http://arxiv.org/abs/2003.10601
AUTHORS:        Md Sirajus Salekin ; Ghada Zamzmi ; Dmitry Goldgof ; Rangachar Kasturi ; Thao Ho ; Yu Sun
COMMENTS:       Accepted in the 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020)
HIGHLIGHT:      This paper presents the first investigation into the use of fully automated deep learning framework for assessing neonatal postoperative pain.

32, TITLE:      A Simple Fix for Convolutional Neural Network via Coordinate Embedding
http://arxiv.org/abs/2003.10589
AUTHORS:        Liliang Ren ; Zhuonan Hao
COMMENTS:       6 pages, 8 figures, Course Project for ECE271B
HIGHLIGHT:      In this project, we proposed a simple approach to incorporate the coordinate information to the CNN model through coordinate embedding.

33, TITLE:      Multifactorial Cellular Genetic Algorithm (MFCGA): Algorithmic Design, Performance Comparison and Genetic Transferability Analysis
http://arxiv.org/abs/2003.10768
AUTHORS:        Eneko Osaba ; Aritz D. Martinez ; Jesus L. Lobo ; Javier Del Ser ; Francisco Herrera
COMMENTS:       Accepted for its presentation at WCCI 2020
HIGHLIGHT:      In this work we propose a novel algorithmic scheme for Multifactorial Optimization scenarios - the Multifactorial Cellular Genetic Algorithm (MFCGA) - that hinges on concepts from Cellular Automata to implement mechanisms for exchanging knowledge among problems.

34, TITLE:      Input representation in recurrent neural networks dynamics
http://arxiv.org/abs/2003.10585
AUTHORS:        Pietro Verzelli ; Cesare Alippi ; Lorenzo Livi ; Peter Tino
HIGHLIGHT:      A novel analysis of the dynamics of such networks is proposed, which allows one to express the state evolution using the controllability matrix.

35, TITLE:      Automated Detection of Cribriform Growth Patterns in Prostate Histology Images
http://arxiv.org/abs/2003.10543
AUTHORS:        Pierre Ambrosini ; Eva Hollemans ; Charlotte F. Kweldam ; Geert J. L. H. van Leenders ; Sjoerd Stallinga ; Frans Vos
COMMENTS:       13 pages, 6 figures
HIGHLIGHT:      We aimed to introduce a deep learning method to detect such patterns automatically.

36, TITLE:      Dynamic Hierarchical Mimicking Towards Consistent Optimization Objectives
http://arxiv.org/abs/2003.10739
AUTHORS:        Duo Li ; Qifeng Chen
COMMENTS:       Accepted by CVPR 2020. Code and pretrained models are available at https://github.com/d-li14/DHM
HIGHLIGHT:      Complementary to previous training strategies, we propose Dynamic Hierarchical Mimicking, a generic feature learning mechanism, to advance CNN training with enhanced generalization ability.

37, TITLE:      FADNet: A Fast and Accurate Network for Disparity Estimation
http://arxiv.org/abs/2003.10758
AUTHORS:        Qiang Wang ; Shaohuai Shi ; Shizhen Zheng ; Kaiyong Zhao ; Xiaowen Chu
HIGHLIGHT:      To this end, we propose an efficient and accurate deep network for disparity estimation named FADNet with three main features: 1) It exploits efficient 2D based correlation layers with stacked blocks to preserve fast computation; 2) It combines the residual structures to make the deeper model easier to learn; 3) It contains multi-scale predictions so as to exploit a multi-scale weight scheduling training technique to improve the accuracy.

38, TITLE:      Scalable learning for bridging the species gap in image-based plant phenotyping
http://arxiv.org/abs/2003.10757
AUTHORS:        Daniel Ward ; Peyman Moghadam
COMMENTS:       Under review. Abstract modified to meed arXiv requirements. Dataset available at: https://danielcward.github.io/UPGen/
HIGHLIGHT:      In this paper, we investigate the use of synthetic data for leaf instance segmentation.

39, TITLE:      Two-Step Surface Damage Detection Scheme using Convolutional Neural Network and Artificial Neural Neural
http://arxiv.org/abs/2003.10760
AUTHORS:        Alice Yi Yang ; Ling Cheng
HIGHLIGHT:      This paper proposes a two-step surface damage detection scheme using Convolutional Neural Network (CNN) and Artificial Neural Network (ANN).

40, TITLE:      Distributional Reinforcement Learning with Ensembles
http://arxiv.org/abs/2003.10903
AUTHORS:        Björn Lindenberg ; Jonas Nordqvist ; Karl-Olof Lindahl
COMMENTS:       15 pages, 2 figures
HIGHLIGHT:      Specifically, we propose an extension to categorical reinforcement learning, where distributional learning targets are implicitly based on the total information gathered by an ensemble.

41, TITLE:      Re-Training StyleGAN -- A First Step Towards Building Large, Scalable Synthetic Facial Datasets
http://arxiv.org/abs/2003.10847
AUTHORS:        Viktor Varkarakis ; Shabab Bazrafkan ; Peter Corcoran
HIGHLIGHT:      In this paper, we recap the StyleGAN architecture and training methodology and present our experiences of retraining it on a number of alternative public datasets.

42, TITLE:      Reservoir Computing with Planar Nanomagnet Arrays
http://arxiv.org/abs/2003.10948
AUTHORS:        Peng Zhou ; Nathan R. McDonald ; Alexander J. Edwards ; Lisa Loomis ; Clare D. Thiem ; Joseph S. Friedman
HIGHLIGHT:      This work proposes a novel hardware implementation of a reservoir computer using a planar nanomagnet array.

43, TITLE:      PanNuke Dataset Extension, Insights and Baselines
http://arxiv.org/abs/2003.10778
AUTHORS:        Jevgenij Gamper ; Navid Alemi Koohbanani ; Simon Graham ; Mostafa Jahanifar ; Syed Ali Khurram ; Ayesha Azam ; Katherine Hewitt ; Nasir Rajpoot
COMMENTS:       Work in progress
HIGHLIGHT:      We study the performance of segmentation and classification models when applied to the proposed dataset and demonstrate the application of models trained on PanNuke to whole-slide images.

44, TITLE:      Organ Segmentation From Full-size CT Images Using Memory-Efficient FCN
http://arxiv.org/abs/2003.10690
AUTHORS:        Chenglong Wang ; Masahiro Oda ; Kensaku Mori
HIGHLIGHT:      In this work, we present a memory-efficient fully convolutional network (FCN) incorporated with several memory-optimized techniques to reduce the run-time GPU memory demand during training phase.

45, TITLE:      Learning to Reconstruct Confocal Microscopy Stacks from Single Light Field Images
http://arxiv.org/abs/2003.11004
AUTHORS:        Josue Page ; Federico Saltarin ; Yury Belyaev ; Ruth Lyck ; Paolo Favaro
COMMENTS:       22 pages, 12 figures
HIGHLIGHT:      We present a novel deep learning approach to reconstruct confocal microscopy stacks from single light field images. To train our network, we built a data set of 362 light field images of mouse brain blood vessels and the corresponding aligned set of 3D confocal scans, which we use as ground truth.

46, TITLE:      Learning regularization and intensity-gradient-based fidelity for single image super resolution
http://arxiv.org/abs/2003.10689
AUTHORS:        Hu Liang ; Shengrong Zhao
HIGHLIGHT:      The regularization-based method can effectively utilize the self-information of observation.

47, TITLE:      Automatic Detection of Coronavirus Disease (COVID-19) Using X-ray Images and Deep Convolutional Neural Networks
http://arxiv.org/abs/2003.10849
AUTHORS:        Ali Narin ; Ceren Kaya ; Ziynet Pamuk
COMMENTS:       The manuscript has 17 pages, 8 figures and 1 table
HIGHLIGHT:      In this study, three different convolutional neural network based models (ResNet50, InceptionV3 and Inception-ResNetV2) have been proposed for the detection of coronavirus pneumonia infected patient using chest X-ray radiographs.

48, TITLE:      TeCNO: Surgical Phase Recognition with Multi-Stage Temporal Convolutional Networks
http://arxiv.org/abs/2003.10751
AUTHORS:        Tobias Czempiel ; Magdalini Paschali ; Matthias Keicher ; Walter Simson ; Hubertus Feussner ; Seong Tae Kim ; Nassir Navab
COMMENTS:       10 pages, 2 figures
HIGHLIGHT:      In this paper, we propose, for the first time in workflow analysis, a Multi-Stage Temporal Convolutional Network (MS-TCN) that performs hierarchical prediction refinement for surgical phase recognition.

49, TITLE:      ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
http://arxiv.org/abs/2003.10555
AUTHORS:        Kevin Clark ; Minh-Thang Luong ; Quoc V. Le ; Christopher D. Manning
COMMENTS:       ICLR 2020
HIGHLIGHT:      As an alternative, we propose a more sample-efficient pre-training task called replaced token detection.

50, TITLE:      Improving Yorùbá Diacritic Restoration
http://arxiv.org/abs/2003.10564
AUTHORS:        Iroro Orife ; David I. Adelani ; Timi Fasubaa ; Victor Williamson ; Wuraola Fisayo Oyewusi ; Olamilekan Wahab ; Kola Tubosun
COMMENTS:       Accepted to ICLR 2020 AfricaNLP workshop
HIGHLIGHT:      We evaluate updated diacritic restoration models on a new, general purpose, public-domain Yor\`ub\'a evaluation dataset of modern journalistic news text, selected to be multi-purpose and reflecting contemporary usage.

51, TITLE:      On the complexity of Broadcast Domination and Multipacking in digraphs
http://arxiv.org/abs/2003.10570
AUTHORS:        Florent Foucaud ; Benjamin Gras ; Anthony Perez ; Florian Sikora
COMMENTS:       Extended abstract accepted in IWOCA 2020
HIGHLIGHT:      We study the complexity of the two dual covering and packing distance-based problems Broadcast Domination and Multipacking in digraphs.

52, TITLE:      Multi-Agent Reinforcement Learning for Problems with Combined Individual and Team Reward
http://arxiv.org/abs/2003.10598
AUTHORS:        Hassam Ullah Sheikh ; Ladislau Bölöni
COMMENTS:       Accepted for publication at International Joint Conference on Neural Networks (IJCNN-2020)
HIGHLIGHT:      To address this problem, we present Decomposed Multi-Agent Deep Deterministic Policy Gradient (DE-MADDPG): a novel cooperative multi-agent reinforcement learning framework that simultaneously learns to maximize the global and local rewards.

53, TITLE:      Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition from a Domain Adaptation Perspective
http://arxiv.org/abs/2003.10780
AUTHORS:        Muhammad Abdullah Jamal ; Matthew Brown ; Ming-Hsuan Yang ; Liqiang Wang ; Boqing Gong
COMMENTS:       Accepted for publication at CVPR2020
HIGHLIGHT:      To this end, we propose to augment the classic class-balanced learning by explicitly estimating the differences between the class-conditioned distributions with a meta-learning approach.

54, TITLE:      EllipBody: A Light-weight and Part-based Representation for Human Pose and Shape Recovery
http://arxiv.org/abs/2003.10873
AUTHORS:        Min Wang ; Feng Qiu ; Wentao Liu ; Chen Qian ; Xiaowei Zhou ; Lizhuang Ma
HIGHLIGHT:      To further improve the efficiency of the task, we propose a light-weight body model called EllipBody, which uses ellipsoids to indicate each body part.

55, TITLE:      Do We Need Depth in State-Of-The-Art Face Authentication?
http://arxiv.org/abs/2003.10895
AUTHORS:        Amir Livne ; Alex Bronstein ; Ron Kimmel ; Ziv Aviv ; Shahaf Grofit
HIGHLIGHT:      Here, we introduce a novel method that learns distinctive geometric features from stereo camera systems without the need to explicitly compute the facial surface or depth map.

56, TITLE:      Dataset Cleaning -- A Cross Validation Methodology for Large Facial Datasets using Face Recognition
http://arxiv.org/abs/2003.10815
AUTHORS:        Viktor Varkarakis ; Peter Corcoran
COMMENTS:       2020 Twelfth International Conference on Quality of Multimedia Experience (QoMEX)
HIGHLIGHT:      In this work, it is presented a semi-automatic method for cleaning the noisy large face datasets with the use of face recognition.

57, TITLE:      Hybrid Classification and Reasoning for Image-based Constraint Solving
http://arxiv.org/abs/2003.11001
AUTHORS:        Maxime Mulamba ; Jayanta Mandi ; Rocsildes Canoy ; Tias Guns
HIGHLIGHT:      In this paper, we explore the hybridization of classifying the images with the reasoning of a constraint solver.

58, TITLE:      Neural Game Engine: Accurate learning ofgeneralizable forward models from pixels
http://arxiv.org/abs/2003.10520
AUTHORS:        Chris Bamford ; Simon Lucas
HIGHLIGHT:      Building upon previous work on the Neural GPU, this paper introduces the Neural Game Engine, as a way to learn models directly from pixels.

59, TITLE:      Learning Object Permanence from Video
http://arxiv.org/abs/2003.10469
AUTHORS:        Aviv Shamsian ; Ofri Kleinfeld ; Amir Globerson ; Gal Chechik
HIGHLIGHT:      Here we introduce the setup of learning Object Permanence from data.

60, TITLE:      Deep Line Art Video Colorization with a Few References
http://arxiv.org/abs/2003.10685
AUTHORS:        Min Shi ; Jia-Qi Zhang ; Shu-Yu Chen ; Lin Gao ; Yu-Kun Lai ; Fang-Lue Zhang
HIGHLIGHT:      In this paper, we propose a deep architecture to automatically color line art videos with the same color style as the given reference images.

61, TITLE:      Gen-LaneNet: A Generalized and Scalable Approach for 3D Lane Detection
http://arxiv.org/abs/2003.10656
AUTHORS:        Yuliang Guo ; Guang Chen ; Peitao Zhao ; Weide Zhang ; Jinghao Miao ; Jingao Wang ; Tae Eun Choe
HIGHLIGHT:      We present a generalized and scalable method, called Gen-LaneNet, to detect 3D lanes from a single image. Moreover, we release a new synthetic dataset and its construction strategy to encourage the development and evaluation of 3D lane detection methods.

62, TITLE:      Modeling Cross-view Interaction Consistency for Paired Egocentric Interaction Recognition
http://arxiv.org/abs/2003.10663
AUTHORS:        Zhongguo Li ; Fan Lyu ; Wei Feng ; Song Wang
COMMENTS:       ICME2020
HIGHLIGHT:      On top of that, we propose to build the relevance between two views using biliear pooling, which capture the consistency of two views in feature-level.

63, TITLE:      CRNet: Cross-Reference Networks for Few-Shot Segmentation
http://arxiv.org/abs/2003.10658
AUTHORS:        Weide Liu ; Chi Zhang ; Guosheng Lin ; Fayao Liu
HIGHLIGHT:      In this paper, we propose a cross-reference network (CRNet) for few-shot segmentation.

64, TITLE:      On Localizing a Camera from a Single Image
http://arxiv.org/abs/2003.10664
AUTHORS:        Pradipta Ghosh ; Xiaochen Liu ; Hang Qiu ; Marcos A. M. Vieira ; Gaurav S. Sukhatme ; Ramesh Govindan
HIGHLIGHT:      In this paper, we explore the following question: under what conditions is it possible to estimate the location of a camera from a single image taken by the camera?

65, TITLE:      Real-time 3D object proposal generation and classification under limited processing resources
http://arxiv.org/abs/2003.10670
AUTHORS:        Xuesong Li ; Jose Guivant ; Subhan Khan
HIGHLIGHT:      To achieve real-time 3D object detection with limited computational resources for robots, we propose an efficient detection method consisting of 3D proposal generation and classification.

66, TITLE:      Robust and On-the-fly Dataset Denoising for Image Classification
http://arxiv.org/abs/2003.10647
AUTHORS:        Jiaming Song ; Lunjia Hu ; Yann Dauphin ; Michael Auli ; Tengyu Ma
HIGHLIGHT:      We address this problem by reasoning counterfactually about the loss distribution of examples with uniform random labels had they were trained with the real examples, and use this information to remove noisy examples from the training set.

67, TITLE:      Tractogram filtering of anatomically non-plausible fibers with geometric deep learning
http://arxiv.org/abs/2003.11013
AUTHORS:        Pietro Astolfi ; Ruben Verhagen ; Laurent Petit ; Emanuele Olivetti ; Jonathan Masci ; Davide Boscaini ; Paolo Avesani
HIGHLIGHT:      In this work, we address the problem of tractogram filtering as a supervised learning problem by exploiting the ground truth annotations obtained with a recent heuristic method, which labels fibers as either anatomically plausible or non-plausible according to well-established anatomical properties.



==========Updates to Previous Papers==========
1, TITLE:       Progressive Cross-camera Soft-label Learning for Semi-supervised Person Re-identification
http://arxiv.org/abs/1908.05669
AUTHORS:        Lei Qi ; Lei Wang ; Jing Huo ; Yinghuan Shi ; Yang Gao
COMMENTS:       Accepted by IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
HIGHLIGHT:      In this paper, we focus on the semi-supervised person re-identification (Re-ID) case, which only has the intra-camera (within-camera) labels but not inter-camera (cross-camera) labels.

2, TITLE:       Collaborative Distillation for Ultra-Resolution Universal Style Transfer
http://arxiv.org/abs/2003.08436
AUTHORS:        Huan Wang ; Yijun Li ; Yuehai Wang ; Haoji Hu ; Ming-Hsuan Yang
COMMENTS:       Accepted by CVPR 2020, higher-resolution images than the camera-ready version
HIGHLIGHT:      In this work, we present a new knowledge distillation method (named Collaborative Distillation) for encoder-decoder based neural style transfer to reduce the convolutional filters.

3, TITLE:       Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection
http://arxiv.org/abs/1912.02424
AUTHORS:        Shifeng Zhang ; Cheng Chi ; Yongqiang Yao ; Zhen Lei ; Stan Z. Li
COMMENTS:       Accepted by CVPR 2020 as Oral; Camera Ready Version
HIGHLIGHT:      In this paper, we first point out that the essential difference between anchor-based and anchor-free detection is actually how to define positive and negative training samples, which leads to the performance gap between them.

4, TITLE:       GraphTCN: Spatio-Temporal Interaction Modeling for Human Trajectory Prediction
http://arxiv.org/abs/2003.07167
AUTHORS:        Chengxin Wang ; Shaofeng Cai ; Gary Tan
COMMENTS:       16 pages, 5 figures, 2 tables
HIGHLIGHT:      To support a more efficient and accurate trajectory prediction, we instead propose a novel CNN-based spatial-temporal graph framework GraphTCN, which captures the spatial and temporal interactions in an input-aware manner.

5, TITLE:       Towards Visually Explaining Variational Autoencoders
http://arxiv.org/abs/1911.07389
AUTHORS:        Wenqian Liu ; Runze Li ; Meng Zheng ; Srikrishna Karanam ; Ziyan Wu ; Bir Bhanu ; Richard J. Radke ; Octavia Camps
COMMENTS:       10 pages, 8 figures, 3 tables, CVPR 2020
HIGHLIGHT:      In this work, we take a step towards bridging this crucial gap, proposing the first technique to visually explain VAEs by means of gradient-based attention.

6, TITLE:       You Only Watch Once: A Unified CNN Architecture for Real-Time Spatiotemporal Action Localization
http://arxiv.org/abs/1911.06644
AUTHORS:        Okan Köpüklü ; Xiangyu Wei ; Gerhard Rigoll
HIGHLIGHT:      In this work, we present YOWO, a unified CNN architecture for real-time spatiotemporal action localization in video streams.

7, TITLE:       FragNet: Writer Identification using Deep Fragment Networks
http://arxiv.org/abs/2003.07212
AUTHORS:        Sheng He ; Lambert Schomaker
HIGHLIGHT:      In this paper, we propose a new benchmark study for writer identification based on word or text block images which approximately contain one word.

8, TITLE:       Action Modifiers: Learning from Adverbs in Instructional Videos
http://arxiv.org/abs/1912.06617
AUTHORS:        Hazel Doughty ; Ivan Laptev ; Walterio Mayol-Cuevas ; Dima Damen
COMMENTS:       CVPR 2020
HIGHLIGHT:      We present a method to learn a representation for adverbs from instructional videos using weak supervision from the accompanying narrations.

9, TITLE:       Learning 3D Part Assembly from a Single Image
http://arxiv.org/abs/2003.09754
AUTHORS:        Yichen Li ; Kaichun Mo ; Lin Shao ; Minhyuk Sung ; Leonidas Guibas
HIGHLIGHT:      Towards this end, we introduce a novel problem, single-image-guided 3D part assembly, along with a learningbased solution.

10, TITLE:      TArC: Incrementally and Semi-Automatically Collecting a Tunisian Arabish Corpus
http://arxiv.org/abs/2003.09520
AUTHORS:        Elisa Gugliotta ; Marco Dinarelli
COMMENTS:       Paper accepted at the Language Resources and Evaluation Conference (LREC) 2020
HIGHLIGHT:      In this article we will describe preliminary work on the TArC semi-automatic construction process and some of the first analyses we developed on TArC.

11, TITLE:      The JDDC Corpus: A Large-Scale Multi-Turn Chinese Dialogue Dataset for E-commerce Customer Service
http://arxiv.org/abs/1911.09969
AUTHORS:        Meng Chen ; Ruixue Liu ; Lei Shen ; Shaozu Yuan ; Jingyan Zhou ; Youzheng Wu ; Xiaodong He ; Bowen Zhou
COMMENTS:       This paper is accepted by LREC 2020 (International Conference on Language Resources and Evaluation )
HIGHLIGHT:      In this paper, we construct a large-scale real scenario Chinese E-commerce conversation corpus, JDDC, with more than 1 million multi-turn dialogues, 20 million utterances, and 150 million words.

12, TITLE:      A Simplified Fully Quantized Transformer for End-to-end Speech Recognition
http://arxiv.org/abs/1911.03604
AUTHORS:        Alex Bie ; Bharat Venkitesh ; Joao Monteiro ; Md. Akmal Haidar ; Mehdi Rezagholizadeh
COMMENTS:       Submitted to IEEE Signal Processing Letters Minor changes in Section 3
HIGHLIGHT:      That being said, in this paper, we work on simplifying and compressing Transformer-based encoder-decoder architectures for the end-to-end ASR task.

13, TITLE:      Pre-trained Models for Natural Language Processing: A Survey
http://arxiv.org/abs/2003.08271
AUTHORS:        Xipeng Qiu ; Tianxiang Sun ; Yige Xu ; Yunfan Shao ; Ning Dai ; Xuanjing Huang
COMMENTS:       Invited Review of Science China Technological Sciences
HIGHLIGHT:      In this survey, we provide a comprehensive review of PTMs for NLP.

14, TITLE:      Context-aware Human Motion Prediction
http://arxiv.org/abs/1904.03419
AUTHORS:        Enric Corona ; Albert Pumarola ; Guillem Alenyà ; Francesc Moreno-Noguer
COMMENTS:       Accepted at CVPR20
HIGHLIGHT:      In this paper, we explore this scenario using a novel context-aware motion prediction architecture.

15, TITLE:      clDice -- a Topology-Preserving Loss Function for Tubular Structure Segmentation
http://arxiv.org/abs/2003.07311
AUTHORS:        Suprosanna Shit ; Johannes C. Paetzold ; Anjany Sekuboyina ; Andrey Zhylka ; Ivan Ezhov ; Alexander Unger ; Josien P. W. Pluim ; Giles Tetteh ; Bjoern H. Menze
COMMENTS:       * The authors Suprosanna Shit and Johannes C. Paetzold contributed equally to the work
HIGHLIGHT:      We introduce a novel similarity measure termed clDice, which is calculated on the intersection of the segmentation masks and their (morphological) skeletons.

16, TITLE:      MutualNet: Adaptive ConvNet via Mutual Learning from Network Width and Resolution
http://arxiv.org/abs/1909.12978
AUTHORS:        Taojiannan Yang ; Sijie Zhu ; Chen Chen ; Shen Yan ; Mi Zhang ; Andrew Willis
HIGHLIGHT:      We propose the width-resolution mutual learning method (MutualNet) to train a network that is executable at dynamic resource constraints to achieve adaptive accuracy-efficiency trade-offs at runtime.

17, TITLE:      Momentum Contrast for Unsupervised Visual Representation Learning
http://arxiv.org/abs/1911.05722
AUTHORS:        Kaiming He ; Haoqi Fan ; Yuxin Wu ; Saining Xie ; Ross Girshick
COMMENTS:       CVPR 2020 camera-ready. Code: https://github.com/facebookresearch/moco
HIGHLIGHT:      We present Momentum Contrast (MoCo) for unsupervised visual representation learning.

18, TITLE:      SNDCNN: Self-normalizing deep CNNs with scaled exponential linear units for speech recognition
http://arxiv.org/abs/1910.01992
AUTHORS:        Zhen Huang ; Tim Ng ; Leo Liu ; Henry Mason ; Xiaodan Zhuang ; Daben Liu
HIGHLIGHT:      Inspired by Self- Normalizing Neural Networks, we propose the self-normalizing deep CNN (SNDCNN) based acoustic model topology, by removing the SC/BN and replacing the typical RELU activations with scaled exponential linear unit (SELU) in ResNet-50.

19, TITLE:      Task-Aware Feature Generation for Zero-Shot Compositional Learning
http://arxiv.org/abs/1906.04854
AUTHORS:        Xin Wang ; Fisher Yu ; Trevor Darrell ; Joseph E. Gonzalez
COMMENTS:       17 pages, 9 figures; substantial content updates with additional experiments
HIGHLIGHT:      In this work, we propose a task-aware feature generation (TFG) framework for compositional learning, which generates features of novel visual concepts by transferring knowledge from previously seen concepts.

20, TITLE:      Self-Assignment Flows for Unsupervised Data Labeling on Graphs
http://arxiv.org/abs/1911.03472
AUTHORS:        Matthias Zisler ; Artjom Zern ; Stefania Petra ; Christoph Schnörr
COMMENTS:       42 pages, 17 figures
HIGHLIGHT:      This paper extends the recently introduced assignment flow approach for supervised image labeling to unsupervised scenarios where no labels are given.

21, TITLE:      MeliusNet: Can Binary Neural Networks Achieve MobileNet-level Accuracy?
http://arxiv.org/abs/2001.05936
AUTHORS:        Joseph Bethge ; Christian Bartz ; Haojin Yang ; Ying Chen ; Christoph Meinel
HIGHLIGHT:      In this paper, we instead present an architectural approach: MeliusNet.

22, TITLE:      Ensemble learning in CNN augmented with fully connected subnetworks
http://arxiv.org/abs/2003.08562
AUTHORS:        Daiki Hirata ; Norikazu Takahashi
COMMENTS:       6 pages, 2 figures, 5 tables
HIGHLIGHT:      In this paper, we propose a new model called EnsNet which is composed of one base CNN and multiple Fully Connected SubNetworks (FCSNs).

23, TITLE:      Efficient two step optimization for large embedded deformation graph based SLAM
http://arxiv.org/abs/1906.08477
AUTHORS:        Jingwei Song ; Fang Bai ; Liang Zhao ; Shoudong Huang ; Rong Xiong
COMMENTS:       This work is accepted by ICRA2020 (2020 International Conference on Robotics and Automation) 7 pages 8 figures
HIGHLIGHT:      In this paper, we propose an approach to decouple nodes of deformation graph in large scale dense deformable SLAM and keep the estimation time to be constant.

24, TITLE:      A Dual Camera System for High Spatiotemporal Resolution Video Acquisition
http://arxiv.org/abs/1909.13051
AUTHORS:        Ming Cheng ; Zhan Ma ; M. Salman Asif ; Yiling Xu ; Haojie Liu ; Wenbo Bao ; Jun Sun
COMMENTS:       To appear in IEEE Transactions on Pattern Analysis and Machine Intelligence
HIGHLIGHT:      We propose an end-to-end learning framework, AWnet, mainly consisting of a FlowNet and a FusionNet that learn an adaptive weighting function in pixel domain to combine inputs in a frame recurrent fashion.

25, TITLE:      Bridge the Domain Gap Between Ultra-wide-field and Traditional Fundus Images via Adversarial Domain Adaptation
http://arxiv.org/abs/2003.10042
AUTHORS:        Lie Ju ; Xin Wang ; Quan Zhou ; Hu Zhu ; Mehrtash Harandi ; Paul Bonnington ; Tom Drummond ; Zongyuan Ge
HIGHLIGHT:      We propose a flexible framework to bridge the domain gap between two domains and co-train a UWF fundus diagnosis model by pseudo-labelling and adversarial learning.

26, TITLE:      Rapid AI Development Cycle for the Coronavirus (COVID-19) Pandemic: Initial Results for Automated Detection & Patient Monitoring using Deep Learning CT Image Analysis
http://arxiv.org/abs/2003.05037
AUTHORS:        Ophir Gozes ; Maayan Frid-Adar ; Hayit Greenspan ; Patrick D. Browning ; Huangqi Zhang ; Wenbin Ji ; Adam Bernheim ; Eliot Siegel
COMMENTS:       19 pages, 6 figures
HIGHLIGHT:      We present a system that utilizes robust 2D and 3D deep learning models, modifying and adapting existing AI models and combining them with clinical understanding.

27, TITLE:      ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
http://arxiv.org/abs/1910.03151
AUTHORS:        Qilong Wang ; Banggu Wu ; Pengfei Zhu ; Peihua Li ; Wangmeng Zuo ; Qinghua Hu
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In particular, we propose an Efficient Channel Attention (ECA) module, which only involves $k (k < 9)$ parameters but brings clear performance gain.

28, TITLE:      Resolution Adaptive Networks for Efficient Inference
http://arxiv.org/abs/2003.07326
AUTHORS:        Le Yang ; Yizeng Han ; Xi Chen ; Shiji Song ; Jifeng Dai ; Gao Huang
COMMENTS:       CVPR 2020
HIGHLIGHT:      In this paper, we focus on spatial redundancy of input samples and propose a novel Resolution Adaptive Network (RANet), which is inspired by the intuition that low-resolution representations are sufficient for classifying "easy" inputs containing large objects with prototypical features, while only some "hard" samples need spatially detailed information.

29, TITLE:      PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimation
http://arxiv.org/abs/1911.04231
AUTHORS:        Yisheng He ; Wei Sun ; Haibin Huang ; Jianran Liu ; Haoqiang Fan ; Jian Sun
COMMENTS:       Accepted to Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2020. (CVPR 2020)
HIGHLIGHT:      In this work, we present a novel data-driven method for robust 6DoF object pose estimation from a single RGBD image.

30, TITLE:      Fully Trainable and Interpretable Non-Local Sparse Models for Image Restoration
http://arxiv.org/abs/1912.02456
AUTHORS:        Bruno Lecouat ; Jean Ponce ; Julien Mairal
HIGHLIGHT:      We propose a novel differentiable relaxation of joint sparsity that exploits both principles and leads to a general framework for image restoration which is (1) trainable end to end, (2) fully interpretable, and (3) much more compact than competing deep learning architectures.

31, TITLE:      Hexagonal Image Processing in the Context of Machine Learning: Conception of a Biologically Inspired Hexagonal Deep Learning Framework
http://arxiv.org/abs/1911.11251
AUTHORS:        Tobias Schlosser ; Michael Friedrich ; Danny Kowerko
COMMENTS:       Accepted for: 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA)
HIGHLIGHT:      This contribution serves as a general application-oriented approach with the synthesis of the therefor designed hexagonal image processing framework, called Hexnet, the processing steps of hexagonal image transformation, and dependent methods.

32, TITLE:      ParasNet: Fast Parasites Detection with Neural Networks
http://arxiv.org/abs/2002.11327
AUTHORS:        X. F. Xu ; S. Talbot ; T. Selvaraja
HIGHLIGHT:      In this work, we studied the bright field based cell level Cryptosporidium and Giardia detection in the drink water with deep learning.

33, TITLE:      FGN: Fusion Glyph Network for Chinese Named Entity Recognition
http://arxiv.org/abs/2001.05272
AUTHORS:        Zhenyu Xuan ; Rui Bao ; Chuyu Ma ; Shengyi Jiang
HIGHLIGHT:      In this paper, we propose the FGN , Fusion Glyph Network for Chinese NER.

34, TITLE:      In-training Matrix Factorization for Parameter-frugal Neural Machine Translation
http://arxiv.org/abs/1910.06393
AUTHORS:        Zachary Kaden ; Teven Le Scao ; Raphael Olivier
HIGHLIGHT:      In this paper, we propose the use of in-training matrix factorization to reduce the model size for neural machine translation.

35, TITLE:      On the Efficiency of the Sinkhorn and Greenkhorn Algorithms and Their Acceleration for Optimal Transport
http://arxiv.org/abs/1906.01437
AUTHORS:        Tianyi Lin ; Nhat Ho ; Michael I. Jordan
COMMENTS:       A preliminary version [arXiv:1901.06482] of this paper, with a subset of the results that are presented here, was presented at ICML 2019
HIGHLIGHT:      We present new complexity results for several algorithms that approximately solve the regularized optimal transport (OT) problem between two discrete probability measures with at most $n$ atoms.

36, TITLE:      TE141K: Artistic Text Benchmark for Text Effect Transfer
http://arxiv.org/abs/1905.03646
AUTHORS:        Shuai Yang ; Wenjing Wang ; Jiaying Liu
COMMENTS:       Accepted by TPAMI 2020. Project page: https://daooshee.github.io/TE141K/
HIGHLIGHT:      To address this problem, we introduce a new text effects dataset, TE141K, with 141,081 text effect/glyph pairs in total.

37, TITLE:      Towards Unsupervised Learning of Generative Models for 3D Controllable Image Synthesis
http://arxiv.org/abs/1912.05237
AUTHORS:        Yiyi Liao ; Katja Schwarz ; Lars Mescheder ; Andreas Geiger
COMMENTS:       CVPR 2020
HIGHLIGHT:      We define the new task of 3D controllable image synthesis and propose an approach for solving it by reasoning both in 3D space and in the 2D image domain.

38, TITLE:      RoIMix: Proposal-Fusion among Multiple Images for Underwater Object Detection
http://arxiv.org/abs/1911.03029
AUTHORS:        Wei-Hong Lin ; Jia-Xing Zhong ; Shan Liu ; Thomas Li ; Ge Li
COMMENTS:       ICASSP 2020
HIGHLIGHT:      We propose an augmentation method called RoIMix, which characterizes interactions among images.

39, TITLE:      GeoGraph: Learning graph-based multi-view object detection with geometric cues end-to-end
http://arxiv.org/abs/2003.10151
AUTHORS:        Ahmed Samy Nassar ; Stefano D'Aronco ; Sébastien Lefèvre ; Jan D. Wegner
HIGHLIGHT:      In this paper we propose an end-to-end learnable approach that detects static urban objects from multiple views, re-identifies instances, and finally assigns a geographic position per object.

40, TITLE:      VIFB: A Visible and Infrared Image Fusion Benchmark
http://arxiv.org/abs/2002.03322
AUTHORS:        Xingchen Zhang ; Ping Ye ; Gang Xiao
COMMENTS:       10 pages, 5 figures, 5 tables
HIGHLIGHT:      In this paper, after briefly reviewing recent advances of visible and infrared image fusion, we present a visible and infrared image fusion benchmark (VIFB) which consists of 21 image pairs, a code library of 20 fusion algorithms and 13 evaluation metrics.

41, TITLE:      Learning Relational Representations with Auto-encoding Logic Programs
http://arxiv.org/abs/1903.12577
AUTHORS:        Sebastijan Dumancic ; Tias Guns ; Wannes Meert ; Hendrik Blockeel
COMMENTS:       8 pages,4 figures, paper + supplement, published at IJCAI
HIGHLIGHT:      This paper introduces a novel framework for relational representation learning that combines the best of both worlds.

42, TITLE:      How the Brain might use Division
http://arxiv.org/abs/2003.05320
AUTHORS:        Kieran Greer
HIGHLIGHT:      In this paper, the author suggests that the maths question can be answered more easily if the problem is changed into one of symbol manipulation and not just number counting.

43, TITLE:      DeepAlign: Alignment-based Process Anomaly Correction using Recurrent Neural Networks
http://arxiv.org/abs/1911.13229
AUTHORS:        Timo Nolle ; Alexander Seeliger ; Nils Thoma ; Max Mühlhäuser
HIGHLIGHT:      In this paper, we propose DeepAlign, a novel approach to multi-perspective process anomaly correction, based on recurrent neural networks and bidirectional beam search.

44, TITLE:      A Comparative Study of Distributional and Symbolic Paradigms for Relational Learning
http://arxiv.org/abs/1806.11391
AUTHORS:        Sebastijan Dumancic ; Alberto Garcia-Duran ; Mathias Niepert
COMMENTS:       corrected version: incorrect evaluation fixed; IJCAI 2019
HIGHLIGHT:      In this work, we compare representation learning and relational learning on various relational classification and clustering tasks and analyse the complexity of the rules used implicitly by these approaches.

45, TITLE:      From Statistical Relational to Neuro-Symbolic Artificial Intelligence
http://arxiv.org/abs/2003.08316
AUTHORS:        Luc De Raedt ; Sebastijan Dumančić ; Robin Manhaeve ; Giuseppe Marra
HIGHLIGHT:      Neuro-symbolic and statistical relational artificial intelligence both integrate frameworks for learning with logical reasoning.

46, TITLE:      Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Prediction
http://arxiv.org/abs/2002.11927
AUTHORS:        Abduallah Mohamed ; Kun Qian ; Mohamed Elhoseiny ; Christian Claudel
COMMENTS:       Accepted by CVPR 2020
HIGHLIGHT:      We propose the Social Spatio-Temporal Graph Convolutional Neural Network (Social-STGCNN), which substitutes the need of aggregation methods by modeling the interactions as a graph.

47, TITLE:      ROAM: Recurrently Optimizing Tracking Model
http://arxiv.org/abs/1907.12006
AUTHORS:        Tianyu Yang ; Pengfei Xu ; Runbo Hu ; Hua Chai ; Antoni B. Chan
COMMENTS:       CVPR2020 camera ready
HIGHLIGHT:      In this paper, we design a tracking model consisting of response generation and bounding box regression, where the first component produces a heat map to indicate the presence of the object at different positions and the second part regresses the relative bounding box shifts to anchors mounted on sliding-window locations.

48, TITLE:      Dedge-AGMNet:an effective stereo matching network optimized by depth edge auxiliary task
http://arxiv.org/abs/1908.09346
AUTHORS:        Weida Yang ; Xindong Ai ; Zuliu Yang ; Yong Xu ; Yong Zhao
HIGHLIGHT:      To improve the performance in ill-posed regions, this paper proposes an atrous granular multi-scale network based on depth edge subnetwork(Dedge-AGMNet).

49, TITLE:      Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling
http://arxiv.org/abs/2003.06060
AUTHORS:        Tong Che ; Ruixiang Zhang ; Jascha Sohl-Dickstein ; Hugo Larochelle ; Liam Paull ; Yuan Cao ; Yoshua Bengio
HIGHLIGHT:      We show that the sum of the implicit generator log-density $\log p_g$ of a GAN with the logit score of the discriminator defines an energy function which yields the true data density when the generator is imperfect but the discriminator is optimal, thus making it possible to improve on the typical generator (with implicit density $p_g$).

50, TITLE:      Deep Reinforcement Learning with Smooth Policy
http://arxiv.org/abs/2003.09534
AUTHORS:        Qianli Shen ; Yan Li ; Haoming Jiang ; Zhaoran Wang ; Tuo Zhao
HIGHLIGHT:      In this paper, we develop a new training framework --- $\textbf{S}$mooth $\textbf{R}$egularized $\textbf{R}$einforcement $\textbf{L}$earning ($\textbf{SR}^2\textbf{L}$), where the policy is trained with smoothness-inducing regularization.

51, TITLE:      Mutual Information Maximization in Graph Neural Networks
http://arxiv.org/abs/1905.08509
AUTHORS:        Xinhan Di ; Pengqian Yu ; Rui Bu ; Mingchao Sun
COMMENTS:       Accepted for presentation at IJCNN 2020
HIGHLIGHT:      We propose a new approach of enlarging the normal neighborhood in the aggregation of GNNs, which aims at maximizing mutual information.

52, TITLE:      On Exact Reznick, Hilbert-Artin and Putinar's Representations
http://arxiv.org/abs/1811.10062
AUTHORS:        Victor Magron ; Mohab Safey El Din
COMMENTS:       35 pages, 4 tables, extended version of the paper from ISSAC'18 conference (available at arXiv::1802.10339)
HIGHLIGHT:      We consider the problem of computing exact sums of squares (SOS) decompositions for certain classes of non-negative multivariate polynomials, relying on semidefinite programming (SDP) solvers.