DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915256459067392 |
|---|---|
| author | Li, Yinqi Chang, Hong Hou, Ruibing Shan, Shiguang Chen, Xilin |
| author_facet | Li, Yinqi Chang, Hong Hou, Ruibing Shan, Shiguang Chen, Xilin |
| contents | Diffusion models have shown remarkable progress in various generative tasks such as image and video generation. This paper studies the problem of leveraging pretrained diffusion models for performing discriminative tasks. Specifically, we extend the discriminative capability of pretrained frozen generative diffusion models from the classification task to the more complex object detection task, by "inverting" a pretrained layout-to-image diffusion model. To this end, a gradient-based discrete optimization approach for replacing the heavy prediction enumeration process, and a prior distribution model for making more accurate use of the Bayes' rule, are proposed respectively. Empirical results show that this method is on par with basic discriminative object detection baselines on COCO dataset. In addition, our method can greatly speed up the previous diffusion-based method for classification without sacrificing accuracy. Code and models are available at https://github.com/LiYinqi/DIVE . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_17253 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks Li, Yinqi Chang, Hong Hou, Ruibing Shan, Shiguang Chen, Xilin Computer Vision and Pattern Recognition Multimedia Diffusion models have shown remarkable progress in various generative tasks such as image and video generation. This paper studies the problem of leveraging pretrained diffusion models for performing discriminative tasks. Specifically, we extend the discriminative capability of pretrained frozen generative diffusion models from the classification task to the more complex object detection task, by "inverting" a pretrained layout-to-image diffusion model. To this end, a gradient-based discrete optimization approach for replacing the heavy prediction enumeration process, and a prior distribution model for making more accurate use of the Bayes' rule, are proposed respectively. Empirical results show that this method is on par with basic discriminative object detection baselines on COCO dataset. In addition, our method can greatly speed up the previous diffusion-based method for classification without sacrificing accuracy. Code and models are available at https://github.com/LiYinqi/DIVE . |
| title | DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks |
| topic | Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2504.17253 |