📚 A curated collection of papers exploring the intersection of
Reinforcement Learning × Computer Vision
This repository is a curated awesome list of research papers at the intersection of Reinforcement Learning (RL) and Computer Vision (CV), sourced from top-tier A* conferences including CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, ICRA, AAAI, and IJCAI.
Papers are organized by CV task to help researchers quickly navigate the rapidly evolving landscape.
Note
RL4CV denotes methods where reinforcement learning is applied to support or improve computer vision tasks and vision pipelines.
Available Tags:
- Highly Cited ⭐
- Benchmarks & Datasets 📊
- 1. Object Detection & Localization
- 2. Image Segmentation
- 3. Image & Video Generation
- 4. Video Understanding & Reasoning
- 5. Image Restoration & Enhancement
- 6. 3D Vision & View Planning
- 7. Medical Image Analysis
- 8. Visual Navigation & Embodied AI
- 9. Vision-Language Models with RL
RL agents sequentially search, zoom, and localize objects in images.
| Paper | Venue | Year | Links |
|---|---|---|---|
| Recurrent Models of Visual Attention ⭐ | NeurIPS | 2014 |  |
| Multiple Object Recognition with Visual Attention ⭐ | ICLR | 2015 | |
| Active Object Localization with Deep Reinforcement Learning ⭐ | ICCV | 2015 | |
| Hierarchical Object Detection with Deep Reinforcement Learning | NeurIPS DRL Workshop | 2016 | |
| Tree-Structured Reinforcement Learning for Sequential Object Localization | NeurIPS | 2016 | |
| Reinforcement Learning for Visual Object Detection ⭐ | CVPR | 2016 | |
| Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization | BMVC | 2016 | |
| Efficient Object Detection in Large Images using Deep Reinforcement Learning | WACV | 2020 | |
| Learning to View: Decision Transformers for Active Object Detection | ICRA | 2023 | |
| Reward Finetuning for Faster and More Accurate Unsupervised Object Discovery | NeurIPS | 2023 | |
| Adaptive Important Region Selection with Reinforced Hierarchical Search for Dense Object Detection | NeurIPS | 2024 | |
RL for interactive, sequential, or region-growing segmentation strategies.
| Paper | Venue | Year | Links |
|---|---|---|---|
| Reinforced Active Learning for Image Segmentation ⭐ | ICLR | 2020 | |
| Embodied Visual Active Learning for Semantic Segmentation | AAAI | 2021 | |
| AlignSAM: Aligning Segment Anything Model to Open Context via Reinforcement Learning | CVPR | 2024 | |
| Focus-Then-Decide: Segmentation-Assisted Reinforcement Learning | AAAI | 2024 | |
| SAM-R1: Leveraging SAM for Reward Feedback in Multimodal Segmentation via Reinforcement Learning | NeurIPS | 2025 | |
Aligning visual generative models with human preferences via RL fine-tuning.
| Paper | Venue | Year | Links |
|---|---|---|---|
| ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation | NeurIPS | 2023 | |
| DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Models | NeurIPS | 2023 | |
| Training Diffusion Models with Reinforcement Learning ⭐ | ICLR | 2024 | |
| RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment | TMLR | 2024 | |
| Rich Human Feedback for Text-to-Image Generation | CVPR | 2024 | |
| Diff-Instruct++: Training One-step Text-to-image Generator Model to Align with Human Preferences | TMLR | 2024 | |
| Training Diffusion Models Towards Diverse Image Generation with Reinforcement Learning | CVPR | 2024 | |
RL for long-horizon temporal reasoning, highlight detection, and video summarization.
| Paper | Venue | Year | Links |
|---|---|---|---|
| End-to-end Learning of Action Detection from Frame Glimpses in Videos | CVPR | 2016 | |
| Reinforced Video Captioning with Entailment Rewards | EMNLP | 2017 | |
| Self-Critical Sequence Training for Image Captioning ⭐ | CVPR | 2017 | |
| Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity-Representativeness Reward ⭐ | AAAI | 2018 | |
| Video Captioning via Hierarchical Reinforcement Learning | CVPR | 2018 | |
| Reinforcement Cutting-Agent Learning for Video Object Segmentation | CVPR | 2018 | |
| AdaFocus: Adaptive Focus for Efficient Video Recognition | ICCV | 2021 | |
| Scaling RL to Long Videos | NeurIPS | 2025 | |
| Video-R1: Reinforcing Video Reasoning in MLLMs | NeurIPS | 2025 | |
RL for adaptive, step-wise image denoising, super-resolution, and retouching.
| Paper | Venue | Year | Links |
|---|---|---|---|
| Deep Reinforcement Learning of Volume-guided Progressive View Inpainting for 3D Point Scene Completion from a Single Depth Image | CVPR (Oral) | 2019 | |
| Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning | CVPR (Spotlight) | 2018 | |
| Fully Convolutional Network with Multi-Step Reinforcement Learning for Image Processing | AAAI | 2019 | |
| Path-Restore: Learning Network Path Selection for Image Restoration | TPAMI | 2021 | |
| MOERL: When Mixture-of-Experts Meet Reinforcement Learning for Adverse Weather Image Restoration | ICCV | 2025 | |
| EditScore: Unlocking Online RL for Image Editing via High-Fidelity Reward Modeling | ICLR | 2026 | |
RL for next-best-view selection, 3D scene reconstruction, and point cloud understanding.
| Paper | Venue | Year | Links |
|---|---|---|---|
| A Reinforcement Learning Approach to the View Planning Problem | CVPR | 2017 | |
| Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks | CVPR | 2018 | |
| Next-Best View Policy for 3D Reconstruction | ECCV Workshop | 2020 | |
| Active 3D Shape Reconstruction from Vision and Touch | NeurIPS | 2021 | |
| GenNBV: Generalizable Next-Best-View Policy for Active 3D Reconstruction | CVPR | 2024 | |
RL for landmark detection, lesion localization, and anatomy-aware segmentation.
| Paper | Venue | Year | Links |
|---|---|---|---|
| Communicative Reinforcement Learning Agents for Landmark Detection in Brain Images | MICCAI MLCN Workshop | 2020 | |
| Iteratively-Refined Interactive 3D Medical Image Segmentation with Multi-Agent Reinforcement Learning ⭐ | CVPR | 2020 | |
| Automatic View Planning with Multi-scale Deep Reinforcement Learning Agents | MICCAI | 2018 | |
| Sequential Attention-based Sampling for Histopathological Analysis | NeurIPS | 2025 | |
RL agents that learn to navigate and reason about 3D environments from visual observations.
| Paper | Venue | Year | Links |
|---|---|---|---|
| Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning ⭐ | ICRA | 2017 | |
| Cognitive Mapping and Planning for Visual Navigation ⭐ | CVPR | 2017 | |
| Embodied Question Answering ⭐ | CVPR | 2018 | |
| Neural Map: Structured Memory for Deep Reinforcement Learning | ICLR | 2018 | |
| Learning to Navigate in Cities Without a Map ⭐ | NeurIPS | 2018 | |
| Visual Semantic Navigation using Scene Priors | ICLR | 2019 | |
| DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Frames ⭐ | ICLR | 2020 | |
Aligning and improving multimodal large language models using reinforcement learning from human or AI feedback.
| Paper | Venue | Year | Links |
|---|---|---|---|
| LLaVA-RLHF: Aligning Large Multimodal Models with Factually Augmented RLHF ⭐ | arXiv | 2023 | |
| RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback ⭐ | CVPR | 2024 | |
| R1-V: Reinforcing Super-Human Visual Perception in Multimodal Large Language Models with Cold-Start RL | arXiv | 2025 | |
| Visual-RFT: Visual Reinforcement Fine-Tuning | arXiv | 2025 | |
⭐ Star this repo if you find it useful!
Made with ❤️ for the CV + RL research community