CVPR 2018 accepted papers on Video
107 of the 979 papers accepted at CVPR 2018 (10.9%) are on Video, compared with 12.9% at CVPR 2017. They fall into 4 topics.
See CVPR 2018 in the interactive map
Submitting to CVPR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Video segmentation and tracking · 56 papers
- Real-World Anomaly Detection in Surveillance Videos (2,159 citations)
- Future Frame Prediction for Anomaly Detection – A New Baseline (1,540 citations)
- Learning Spatial-Temporal Regularized Correlation Filters for Visual Tracking (891 citations)
- Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (867 citations)
- Neural Sign Language Translation (787 citations)
- Deep Video Super-Resolution Network Using Dynamic Upsampling Filters Without Explicit Motion Compensation (636 citations)
- Frame-Recurrent Video Super-Resolution (602 citations)
- Context-Aware Synthesis for Video Frame Interpolation (435 citations)
- Efficient Video Object Segmentation via Network Modulation (389 citations)
- Diversity Regularized Spatiotemporal Attention for Video-Based Person Re-Identification (361 citations)
- Convolutional Sequence to Sequence Model for Human Dynamics (360 citations)
- End-to-End Flow Correlation Tracking With Spatial-Temporal Attention (314 citations)
and 44 more in the interactive map.
Action recognition · 35 papers
- A Closer Look at Spatiotemporal Convolutions for Action Recognition (3,632 citations)
- Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet? (2,278 citations)
- AVA: A Video Dataset of Spatio-Temporally Localized Atomic Visual Actions (1,021 citations)
- Rethinking the Faster R-CNN Architecture for Temporal Action Localization (781 citations)
- Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition (466 citations)
- Weakly Supervised Action Localization by Sparse Temporal Pooling Network (429 citations)
- Appearance-and-Relation Networks for Video Classification (424 citations)
- Compressed Video Action Recognition (369 citations)
- PoTion: Pose MoTion Representation for Action Recognition (351 citations)
- Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognition (349 citations)
- Learning and Using the Arrow of Time (347 citations)
- MiCT: Mixed 3D/2D Convolutional Tube for Human Action Recognition (257 citations)
and 23 more in the interactive map.
Video-language understanding · 13 papers
- End-to-End Dense Video Captioning With Masked Transformer (578 citations)
- Reconstruction Network for Video Captioning (340 citations)
- Video Captioning via Hierarchical Reinforcement Learning (274 citations)
- Motion-Appearance Co-Memory Networks for Video Question Answering (261 citations)
- Bidirectional Attentive Fusion With Context Gating for Dense Video Captioning (224 citations)
- Jointly Localizing and Describing Events for Dense Video Captioning (180 citations)
- M3: Multimodal Memory Modelling for Video Captioning (168 citations)
- Attend and Interact: Higher-Order Object Interactions for Video Understanding (144 citations)
- MovieGraphs: Towards Understanding Human-Centric Situations From Videos (136 citations)
- What Makes a Video a Video: Analyzing Temporal Information in Video Understanding Models and Datasets (132 citations)
- Finding "It": Weakly-Supervised Reference-Aware Visual Grounding in Instructional Videos (99 citations)
- Interpretable Video Captioning via Trajectory Structured Localization (66 citations)
and 1 more in the interactive map.
Video generation · 3 papers
- MoCoGAN: Decomposing Motion and Content for Video Generation (1,116 citations)
- Controllable Video Generation With Sparse Trajectories (96 citations)
- Every Smile Is Unique: Landmark-Guided Diverse Smile Generation (60 citations)
