DRL for Video Analysis
Object (face) Detection, Tracking, and Recognition
- Deep Reinforcement Learning with Iterative Shift for Visual Tracking [Link]
- Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning [Link]
- Tracking as Online Decision-Making: Learning a Policy from Streaming Videos with Reinforcement Learning [Link]
- Dual-Agent Deep Reinforcement Learning for Deformable Face Tracking [Link]
- Collaborative Deep Reinforcement Learning for Multi-object Tracking [Link]
- Attention-aware deep reinforcement learning for video face recognition [Link]
- Hyperparameter Optimization for Tracking With Continuous Deep Q-Learning [Link]
Action Detection, Recognition, and Prediction
- Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition [Link]
- Part-Activated Deep Reinforcement Learning for Action Prediction [Link]
- A Self-Adaptive Proposal Model for Temporal Action Detection based on Reinforcement Learning [Link]
Video Summary and Caption
- FFNet: Video fast-forwarding via reinforcement learning [Link]
- Video captioning via hierarchical reinforcement learning [Link]
DRL for Network Structure Learning
- Neural architecture search with reinforcement learning [Link]
- Learning transferable architectures for scalable image recognition [Link]
- AMC: AutoML for model compression and acceleration on mobile devices [Link]
- HAQ: Hardware-aware Automated Quantization with Mixed-precision [Link]
- Runtime neural pruning [Link]
- Runtime Network Routing for Efficient Image Classification [Link]
- Skipnet: Learning dynamic routing in convolutional networks [Link]
- Dynamic Progressive Pruning for Efficient Video Classification
DRL for Image Editing & Understanding
Object Detection/Localization
- Active object localization with deep reinforcement learning [Link]
- Deep Reinforcement Learning of Region Proposal Networks for Object Detection [Link]
- Too Far to See? Not Really!—Pedestrian Detection With Scale-Aware Localization Policy [Link]
- Learning globally optimized object detector via policy gradient [Link]
- Collaborative deep reinforcement learning for joint object search [Link]
- Hierarchical Object Detection with Deep Reinforcement Learning [Link]
- Tree-Structured Reinforcement Learning for Sequential Object Localization [Link]
Image Editing/Enhancement
- A2-RL: aesthetics aware reinforcement learning for image cropping [Link]
- Distort-and-recover: Color enhancement using deep reinforcement learning [Link]
- Attention-aware face hallucination via deep reinforcement learning [Link]
- Deep variation-structured reinforcement learning for visual relationship and attribute detection [Link]
Visual QA
- Learning Cooperative Visual Dialog Agents with Deep Reinforcement Learning [Link]
- Inverse Visual Question Answering: A New Benchmark and VQA Diagnosis Tool [Link]
Image Captioning
- Deep Reinforcement Learning-based Image Captioning with Embedding Reward [Link]
Other
- 3DCNN-DQN-RNN: a deep reinforcement learning framework for semantic parsing of large-scale 3D point clouds
- GraphBit: Bitwise Interaction Mining via Deep Reinforcement Learning
- Reinforcement Cutting-Agent Learning for Video Object Segmentation
- CIRL: Controllable imitative reinforcement learning for vision-based self-driving
- Sidekick Policy Learning for Active Visual Exploration
- Relaxation-Free Deep Hashing via Policy Gradient
- R2P2: A reparameterized pushforward policy for diverse, precise generative path forecasting
- Real-Time ‘Actor-Critic’Tracking
- First-Person Activity Forecasting from Video with Online Inverse Reinforcement Learning
[B] Action Detection End-to-end learning of action detection from frame glimpses in videos
[C] Visual Tracking Deep Reinforcement Learning for Visual Object Tracking in Videos Learning to track: Online multi-object tracking by decision making
[D] Pose-Estimation and View-Planning Problem PoseAgent: Budget-Constrained 6D Object Pose Estimation via Reinforcement Learning A Reinforcement Learning Approach to the View Planning Problem
References: