Research map / Video / Video-language understanding
EgoEdit: Dataset, Real-Time Streaming Model, and Benchmark for Egocentric Video Editing
Runjia Li, Moayed Haji-Ali, Ashkan Mirzaei, Chaoyang Wang, Arpit Sahni, Ivan Skorokhodov, Aliaksandr Siarohin, Tomas Jakab, Junlin Han, Sergey Tulyakov, Philip Torr, Willi Menapace
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 |
|---|---|---|---|
| add EgoEditData (0% to 25%) (ablation) | EgoEdit · EgoEditBench (10k-iteration checkpoint) | VLM score: 4.87 -> 7.12 | Improved |
| increase EgoEditData from 25% to 75% (ablation) | EgoEdit · EgoEditBench (10k-iteration checkpoint) | VLM score: 7.12 -> 7.52 | Improved |
| increase EgoEditData from 75% to 100% (ablation) | EgoEdit · EgoEditBench (10k-iteration checkpoint) | VLM score: 7.52 -> 7.85 | Improved |
| add 4-step DMD distillation (ablation) | EgoEdit · EgoEditBench | VLM score: 7.76 -> 7.31 | Hurt |
| add 4-step DMD distillation (ablation) | EgoEdit · EgoEditBench | first chunk latency (ms): 13432 -> 6925 | Improved |
| add Self Forcing after DMD (ablation) | EgoEdit · EgoEditBench | VLM score: 7.31 -> 7.71 | Improved |
| add Self Forcing after DMD (ablation) | EgoEdit · EgoEditBench | first chunk latency (ms): 6925 -> 855 | Improved |
| distill with DMD plus Self Forcing vs no distillation (ablation) | EgoEdit · EgoEditBench | VLM score: 7.76 -> 7.71 | No clear effect |
| use EgoEdit-RT instead of StreamDiffusionV2 | streaming editor · EgoEditBench | VLM score: 2.55 -> 7.71 | Improved |
| use EgoEdit instead of Lucy Edit | instruction-guided editor · EgoEditBench | VLM score: 5.44 -> 7.76 | Improved |
| shift from general editing to egocentric tasks | EgoEdit · EgoEditBench vs EditVerseBench | VLM score: 8.00 -> 7.76 | Hurt |
| shift from general editing to egocentric tasks | Lucy Edit · EgoEditBench vs EditVerseBench | VLM score: 6.27 -> 5.44 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- EditVerse — EditVerse: Unifying Image and Video Editing and Generation with In-Context Learning (ICLR 2026) · better on 0 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
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.89)
- Object-Shot Enhanced Grounding Network for Egocentric Video (CVPR 2025 · similarity 0.87)
- HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing (NeurIPS 2026 · similarity 0.87)
- Retrieval-Augmented Egocentric Video Captioning (CVPR 2024 · similarity 0.87)
- Test-Time Adaptation for Combating Missing Modalities in Egocentric Videos (ICLR 2025 · similarity 0.86)
- EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric View of Procedural Activities in Real World (CVPR 2024 · similarity 0.86)
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.
