Holistic Order Prediction in Natural Scenes
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_ | 1866914113474527232 |
|---|---|
| author | Musacchio, Pierre Lee, Hyunmin Park, Jaesik |
| author_facet | Musacchio, Pierre Lee, Hyunmin Park, Jaesik |
| contents | Even in controlled settings, understanding instance-wise geometries is a challenging task for a wide range of visual models. Although specialized systems exist, modern arts rely on expensive input formats (category labels, binary segmentation masks) and inference costs (a quadratic amount of forward passes). We mitigate these limitations by proposing InstaFormer, a network capable of holistic order prediction. That is, solely given an input RGB image, InstaFormer returns the full occlusion and depth orderings for all the instances in the scene in a single forward pass. At its core, InstaFormer relies on interactions between object queries and latent mask descriptors that semantically represent the same objects while carrying complementary information. We comprehensively benchmark and ablate our approach to highlight its effectiveness. Our code and models are open-source and available at this URL: https://github.com/SNU-VGILab/InstaOrder. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_01704 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Holistic Order Prediction in Natural Scenes Musacchio, Pierre Lee, Hyunmin Park, Jaesik Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Even in controlled settings, understanding instance-wise geometries is a challenging task for a wide range of visual models. Although specialized systems exist, modern arts rely on expensive input formats (category labels, binary segmentation masks) and inference costs (a quadratic amount of forward passes). We mitigate these limitations by proposing InstaFormer, a network capable of holistic order prediction. That is, solely given an input RGB image, InstaFormer returns the full occlusion and depth orderings for all the instances in the scene in a single forward pass. At its core, InstaFormer relies on interactions between object queries and latent mask descriptors that semantically represent the same objects while carrying complementary information. We comprehensively benchmark and ablate our approach to highlight its effectiveness. Our code and models are open-source and available at this URL: https://github.com/SNU-VGILab/InstaOrder. |
| title | Holistic Order Prediction in Natural Scenes |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2510.01704 |