Research map / Video / Action recognition
Actor and Observer: Joint Modeling of First and Third-Person Videos
Gunnar A. Sigurdsson, Abhinav Gupta, Cordelia Schmid, Ali Farhadi, Karteek Alahari
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use learned frame weights to select the 10% highest-weight test examples | ActorObserverNet · Charades-Ego, same person | correspondence classification accuracy (%): 50.0 -> 97.2 | Improved |
| Use learned frame weights to select the 10% highest-weight test examples | ActorObserverNet · Charades-Ego, different persons | correspondence classification accuracy (%): 50.0 -> 98.8 | Improved |
| Jointly learn a first/third-person representation | ActorObserverNet · Charades-Ego, same person | median alignment error (seconds): 6.5 -> 5.2 | Improved |
| Jointly learn a first/third-person representation | ActorObserverNet · Charades-Ego, different persons | median alignment error (seconds): 7.0 -> 6.1 | Improved |
| Transfer third-person action knowledge using the joint representation | ActorObserverNet · Charades-Ego test, zero-shot first-person action recognition | video-level mAP: 22.7 -> 25.9 | Improved |
| Share parameters between the two first-person fully connected streams (ablation) | ActorObserverNet · Mapping third-person to first-person | correspondence classification accuracy | No clear effect |
| Share parameters between the two first-person fully connected streams (ablation) | ActorObserverNet · Alignment and localization; zero-shot action recognition | performance | Improved |
| Constrain the Tanh selector scale to be positive (ablation) | ActorObserverNet · Alignment and localization; zero-shot action recognition | performance | Improved |
Closest papers by meaning
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs (NeurIPS 2025 · similarity 0.90)
- Retrieval-Augmented Egocentric Video Captioning (CVPR 2024 · similarity 0.89)
- Action Scene Graphs for Long-Form Understanding of Egocentric Videos (CVPR 2024 · similarity 0.88)
- Bootstrap Your Own Views: Masked Ego-Exo Modeling for Fine-grained View-invariant Video Representations (CVPR 2025 · similarity 0.88)
- Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives (CVPR 2024 · similarity 0.88)
- Viewpoint Rosetta Stone: Unlocking Unpaired Ego-Exo Videos for View-invariant Representation Learning (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
