CVPR 2022 accepted papers by research area
2,077 papers were accepted at CVPR 2022 (+416 vs 2021). Here is how they split across 20 research areas, and which topics gained the most ground.
See CVPR 2022 in the interactive map
Submitting to CVPR? Checklist →
Research areas
| Area | Papers | Share | Change in share vs 2021 |
|---|---|---|---|
| 3D vision | 370 | 17.8% | steady |
| Label-efficient and robust learning | 337 | 16.2% | +0.2 pts |
| Detection and segmentation | 337 | 16.2% | -1.2 pts |
| Video | 222 | 10.7% | +0.5 pts |
| Image generation | 194 | 9.3% | -2.5 pts |
| Architectures and efficiency | 137 | 6.6% | +0.6 pts |
| Vision-language and multimodal | 119 | 5.7% | +0.7 pts |
| Point clouds | 98 | 4.7% | -0.4 pts |
| Adversarial robustness and security | 63 | 3.0% | +0.3 pts |
| Graph learning | 40 | 1.9% | steady |
| Continual learning | 37 | 1.8% | +0.3 pts |
| Probabilistic and scientific modeling | 34 | 1.6% | +0.4 pts |
| Learning theory and algorithms | 27 | 1.3% | +0.2 pts |
| Federated learning and privacy | 24 | 1.2% | +0.6 pts |
| Multi-agent systems and games | 13 | 0.6% | +0.2 pts |
| Optimization | 11 | 0.5% | steady |
| Large language models | 7 | 0.3% | steady |
| Reinforcement learning | 4 | 0.2% | steady |
| Time series | 2 | 0.1% | steady |
| Bandits and online learning | 1 | 0.0% | steady |
Fastest-growing topics at CVPR 2022
Change in share of CVPR papers from 2021 to 2022, topics with at least 10 papers.
- Attention and transformers (Architectures and efficiency): 5 → 55 papers
- NeRF and radiance fields (3D vision): 17 → 54 papers
- Self-supervised and contrastive (Label-efficient and robust learning): 21 → 46 papers
- CLIP and zero-shot (Vision-language and multimodal): 1 → 20 papers
- 3D generation (3D vision): 31 → 54 papers
- Weakly supervised detection and segmentation (Detection and segmentation): 32 → 55 papers
- Zero-/few-shot and meta-learning (Label-efficient and robust learning): 16 → 29 papers
- Variational and generative sampling (Probabilistic and scientific modeling): 2 → 11 papers
- Personalized and heterogeneous FL (Federated learning and privacy): 3 → 12 papers
- Autoregressive and latent generation (Image generation): 24 → 38 papers
More venues
Same year: ICML 2022 · NeurIPS 2022 · ICLR 2022 · AAAI 2022 · ECCV 2022 · TPAMI 2022 · IJCV 2022 · JMLR 2022
CVPR in other years: 2016 · 2017 · 2018 · 2019 · 2020 · 2021 · 2023 · 2024 · 2025 · 2026
