Paper!!!

code+paper:

• Gitxiv
• arxiv

CNN:

• Wide Residual Networks (code)

• Deep Residual Learning for Image Recognition (github) (CVPR 2016)

Semantic Segmentation:

• Instance-Level Segmentation with Deep Densely Connected MRFs

• Conditional Random Fields as Recurrent Neural Networks

• Fully Convolutional Networks for Semantic Segmentation

• Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding

• PARSENET: LOOKING WIDER TO SEE BETTER(ICLR 2016)

• MULTI-SCALE CONTEXT AGGREGATION BY DILATED CONVOLUTIONS (ICLR 2016) (github)

• DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (code)

Optical Flow:

• Fast Optical Flow using Dense Inverse Search (code)

Region proposal:

• AttractioNet (code)

Detection:

• Putting Objects in Perspective(CVPR2006)

• 3D Object Proposals for Accurate Object Class Detection(CVPR2015)

• Fast R-CNN

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

• A MultiPath Network for Object Detection(FAIR)(MSCOCO 2015 2nd)

• LocNet: Improving Localization Accuracy for Object Detection(CVPR 2016 oral)

• HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection(CVPR 2016 spotlight oral)

• Training Region-based Object Detectors with Online Hard Example Mining(CVPR 2016 oral)

• R-FCN: Object Detection via Region-based Fully Convolutional Networks

• SSD: Single Shot MultiBox Detector (ECCV 2016)

• Vehicle Detection from 3D Lidar Using Fully Convolutional Network

Tracking:

• Learning to Track: Online Multi-Object Tracking by Decision Making

• Hierarchical Convolutional Features for Visual Tracking

• Tracking-Learning-Detection

Scene:

• Detecting Vanishing Points using Global Image Context in a Non-Manhattan World (CVPR2016)

• Probabilistic Models for 3D Urban Scene Understanding

3D Localization:

• Joint SFM and Detection Cues for Monocular 3D Localization in Road Scenes(CVPR2015)

Recurrent Neural Networks:

• Recurrent Neural Networks for Driver Activity Anticipation via Sensory-Fusion Architecture

• Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank(NLP)

• Social LSTM: Human Trajectory Prediction in Crowded Spaces (CVPR2016)

• Detecting events and key actors in multi-person videos (CVPR2016)

• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting (NIPS 2015)

• Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences (NIPS 2016)

• Using Fast Weights to Attend to the Recent Past (NIPS 2016)

• [Social Scene Understanding: End-to-End Multi-Person Action Localization and Collective Activity Recognition] (https://arxiv.org/pdf/1611.09078v1.pdf)(arXiv:1611.09078v1 28 Nov 2016)

Action Recognition

• VideoLSTM Convolves, Attends and Flows for Action Recognition

• Online Action Detection(ECCV 2016)

• Temporal Segment Networks: Towards Good Practices for Deep Action Recognition (ECCV 2016)

Self-driving：

• Car that Knows Before You Do: Anticipating Maneuvers via Learning Temporal Driving Models(ICCV2015)

• Learning image representations tied to ego-motion

• Monocular 3D Object Detection for Autonomous Driving (CVPR 2016)

• MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving(arXiv:1612.07695v1 22 Dec 2016)

Depth Estimation:

• Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields(CVPR 2015)

• Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue(ECCV 2016)

Attention

• Top-down Neural Attention by Excitation Backprop(ECCV 2016 oral)

• Dual Attention Networks for Multimodal Reasoning and Matching

Cool

• Deep3D: Fully Automatic 2D-to-3D Video Conversion with Deep Convolutional Neural Networks

3D object shape

• Category-Specific Object Reconstruction from a Single Image

Video Recognition

• Deep Feature Flow for Video Recognition