Self-supervised Pretraining and Finetuning for Monocular Depth and Visual Odometry
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909225209298944 |
|---|---|
| author | Chidlovskii, Boris Antsfeld, Leonid |
| author_facet | Chidlovskii, Boris Antsfeld, Leonid |
| contents | For the task of simultaneous monocular depth and visual odometry estimation, we propose learning self-supervised transformer-based models in two steps. Our first step consists in a generic pretraining to learn 3D geometry, using cross-view completion objective (CroCo), followed by self-supervised finetuning on non-annotated videos. We show that our self-supervised models can reach state-of-the-art performance 'without bells and whistles' using standard components such as visual transformers, dense prediction transformers and adapters. We demonstrate the effectiveness of our proposed method by running evaluations on six benchmark datasets, both static and dynamic, indoor and outdoor, with synthetic and real images. For all datasets, our method outperforms state-of-the-art methods, in particular for depth prediction task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_11019 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Self-supervised Pretraining and Finetuning for Monocular Depth and Visual Odometry Chidlovskii, Boris Antsfeld, Leonid Computer Vision and Pattern Recognition For the task of simultaneous monocular depth and visual odometry estimation, we propose learning self-supervised transformer-based models in two steps. Our first step consists in a generic pretraining to learn 3D geometry, using cross-view completion objective (CroCo), followed by self-supervised finetuning on non-annotated videos. We show that our self-supervised models can reach state-of-the-art performance 'without bells and whistles' using standard components such as visual transformers, dense prediction transformers and adapters. We demonstrate the effectiveness of our proposed method by running evaluations on six benchmark datasets, both static and dynamic, indoor and outdoor, with synthetic and real images. For all datasets, our method outperforms state-of-the-art methods, in particular for depth prediction task. |
| title | Self-supervised Pretraining and Finetuning for Monocular Depth and Visual Odometry |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.11019 |