Unique Lives, Shared World: Learning from Single-Life Videos
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914603891425280 |
|---|---|
| author | Han, Tengda Ebrahimi, Sayna Gokay, Dilara Ku, Li Yang Ovsjanikov, Maks Babukova, Iva Zoran, Daniel Patraucean, Viorica Carreira, Joao Zisserman, Andrew Damen, Dima |
| author_facet | Han, Tengda Ebrahimi, Sayna Gokay, Dilara Ku, Li Yang Ovsjanikov, Maks Babukova, Iva Zoran, Daniel Patraucean, Viorica Carreira, Joao Zisserman, Andrew Damen, Dima |
| contents | We introduce the "single-life" learning paradigm, where we train a distinct vision model exclusively on egocentric videos captured by one individual. We leverage the multiple viewpoints naturally captured within a single life to learn a visual encoder in a self-supervised manner. Our experiments demonstrate three key findings. First, models trained independently on different lives develop a highly aligned geometric understanding. We demonstrate this by training visual encoders on distinct datasets each capturing a different life, both indoors and outdoors, as well as introducing a novel cross-attention-based metric to quantify the functional alignment of the internal representations developed by different models. Second, we show that single-life models learn generalizable geometric representations that effectively transfer to downstream tasks, such as depth estimation, in unseen environments. Third, we demonstrate that training on up to 30 hours from one week of the same person's life leads to comparable performance to training on 30 hours of diverse web data, highlighting the strength of single-life representation learning. Overall, our results establish that the shared structure of the world, both leads to consistency in models trained on individual lives, and provides a powerful signal for visual representation learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_04085 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Unique Lives, Shared World: Learning from Single-Life Videos Han, Tengda Ebrahimi, Sayna Gokay, Dilara Ku, Li Yang Ovsjanikov, Maks Babukova, Iva Zoran, Daniel Patraucean, Viorica Carreira, Joao Zisserman, Andrew Damen, Dima Computer Vision and Pattern Recognition We introduce the "single-life" learning paradigm, where we train a distinct vision model exclusively on egocentric videos captured by one individual. We leverage the multiple viewpoints naturally captured within a single life to learn a visual encoder in a self-supervised manner. Our experiments demonstrate three key findings. First, models trained independently on different lives develop a highly aligned geometric understanding. We demonstrate this by training visual encoders on distinct datasets each capturing a different life, both indoors and outdoors, as well as introducing a novel cross-attention-based metric to quantify the functional alignment of the internal representations developed by different models. Second, we show that single-life models learn generalizable geometric representations that effectively transfer to downstream tasks, such as depth estimation, in unseen environments. Third, we demonstrate that training on up to 30 hours from one week of the same person's life leads to comparable performance to training on 30 hours of diverse web data, highlighting the strength of single-life representation learning. Overall, our results establish that the shared structure of the world, both leads to consistency in models trained on individual lives, and provides a powerful signal for visual representation learning. |
| title | Unique Lives, Shared World: Learning from Single-Life Videos |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.04085 |