Visual Accommodation: Rethinking Image Scale as a Learnable Variable for Object Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914561935802368 |
|---|---|
| author | Seo, Daeun Yang, Hoeseok Park, Sihyeong Kim, Hyungshin |
| author_facet | Seo, Daeun Yang, Hoeseok Park, Sihyeong Kim, Hyungshin |
| contents | We propose Ciliary-DETR (previous name: Elastic-DETR), a framework for test-time resolution adjustment analogous to biological accommodation. While multi-scale data augmentation improves robustness to scale variation, modern detectors rely on fixed inference resolutions, potentially limiting flexibility and robustness. Similar to the ciliary muscle, we introduce a lightweight scale predictor that dynamically estimates test-time scale factors across a wide range of input scales. The core challenge is that the optimal input scale is inherently unobservable under standard training setups. To address this challenge, we introduce a parametric formulation of desired scaling behavior, leading to loss-driven objectives that guide scale optimization. Overall, our method enables flexible and efficient single-pass inference, bridging the gap between training-time robustness and test-time adaptation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_06341 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Visual Accommodation: Rethinking Image Scale as a Learnable Variable for Object Detection Seo, Daeun Yang, Hoeseok Park, Sihyeong Kim, Hyungshin Computer Vision and Pattern Recognition Artificial Intelligence We propose Ciliary-DETR (previous name: Elastic-DETR), a framework for test-time resolution adjustment analogous to biological accommodation. While multi-scale data augmentation improves robustness to scale variation, modern detectors rely on fixed inference resolutions, potentially limiting flexibility and robustness. Similar to the ciliary muscle, we introduce a lightweight scale predictor that dynamically estimates test-time scale factors across a wide range of input scales. The core challenge is that the optimal input scale is inherently unobservable under standard training setups. To address this challenge, we introduce a parametric formulation of desired scaling behavior, leading to loss-driven objectives that guide scale optimization. Overall, our method enables flexible and efficient single-pass inference, bridging the gap between training-time robustness and test-time adaptation. |
| title | Visual Accommodation: Rethinking Image Scale as a Learnable Variable for Object Detection |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2412.06341 |