Research map / Video / Video segmentation and tracking
Ego-Exo: Transferring Visual Representations From Third-Person to First-Person Videos
Yanghao Li, Tushar Nagarajan, Bo Xiong, Kristen Grauman
Check 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 |
|---|---|---|---|
| Add Ego-Score, Object-Score, and Interaction-Map distillation to Kinetics pre-training | Slow-only ResNet-50 · Charades-Ego | mAP: 24.69 -> 26.23 | Improved |
| Add Ego-Score, Object-Score, and Interaction-Map distillation to Kinetics pre-training | Slow-only ResNet-50 · EPIC-Kitchens-100 verbs | top-1 accuracy: +1.64% | Improved |
| Add Ego-Score, Object-Score, and Interaction-Map distillation to Kinetics pre-training | Slow-only ResNet-50 · EPIC-Kitchens-100 nouns | top-1 accuracy: +1.97% | Improved |
| Add Ego-Score distillation (ablation) | Ego-Exo pre-training · EPIC-Kitchens-100 | top-1 accuracy | Improved |
| Add Object-Score distillation (ablation) | Ego-Exo pre-training · Charades-Ego and EPIC-Kitchens-100 | mAP and accuracy | Improved |
| Add Interaction-Map distillation (ablation) | Ego-Exo pre-training · Charades-Ego and EPIC-Kitchens-100 | mAP and accuracy | Improved |
| Add Interaction-Map loss during fine-tuning (ablation) | Third-only and Ego-Exo · EPIC-Kitchens-100 nouns top-5 | top-5 accuracy: 69.72 -> 68.33; 70.28 -> 70.68 | Mixed |
| Use both hand-map and object-map rather than hand-map alone (ablation) | Interaction-Map pre-training · EPIC-Kitchens-100 verbs | top-1 accuracy: 61.35 -> 62.55 | Improved |
| Use both hand-map and object-map rather than object-map alone (ablation) | Interaction-Map pre-training · Charades-Ego | mAP: 26.15 -> 25.91 | Hurt |
| Distill auxiliary-task predictions instead of using auxiliary models directly as pre-trained backbones (ablation) | Ego-Exo · Charades-Ego | mAP: 23.29 -> 25.01; 22.17 -> 25.49 | Improved |
| Distill auxiliary-task predictions instead of appending auxiliary model embeddings (ablation) | Ego-Exo · Charades-Ego | mAP: 24.92 -> 26.23; 24.73 -> 26.23; 24.68 -> 26.23 | Improved |
| Remove implicit pairing information from training the Ego-Score teacher (ablation) | Ego-Exo · Charades-Ego | mAP: 26.23 -> 26.29 | No clear effect |
Later papers that compared with this method
3 later accepted papers report EgoExo as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Sound Bridge: Associating Egocentric and Exocentric Videos via Audio Cues (CVPR 2025 · better on 6 of 6 shared results)
- Learning from Semantic Alignment between Unpaired Multiviews for Egocentric Video Recognition (ICCV 2023 · better on 1 of 2 shared results)
- Egocentric Video-Language Pretraining (NeurIPS 2022 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Retrieval-Augmented Egocentric Video Captioning (CVPR 2024 · similarity 0.91)
- Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives (CVPR 2024 · similarity 0.89)
- Bootstrap Your Own Views: Masked Ego-Exo Modeling for Fine-grained View-invariant Video Representations (CVPR 2025 · similarity 0.89)
- Exocentric-to-Egocentric Video Generation (NeurIPS 2024 · similarity 0.89)
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.88)
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs (NeurIPS 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.
