Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2303.14389 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914687125291008 |
|---|---|
| author | Gao, Shanghua Zhou, Pan Cheng, Ming-Ming Yan, Shuicheng |
| author_facet | Gao, Shanghua Zhou, Pan Cheng, Ming-Ming Yan, Shuicheng |
| contents | Despite its success in image synthesis, we observe that diffusion probabilistic models (DPMs) often lack contextual reasoning ability to learn the relations among object parts in an image, leading to a slow learning process. To solve this issue, we propose a Masked Diffusion Transformer (MDT) that introduces a mask latent modeling scheme to explicitly enhance the DPMs' ability to contextual relation learning among object semantic parts in an image. During training, MDT operates in the latent space to mask certain tokens. Then, an asymmetric diffusion transformer is designed to predict masked tokens from unmasked ones while maintaining the diffusion generation process. Our MDT can reconstruct the full information of an image from its incomplete contextual input, thus enabling it to learn the associated relations among image tokens. We further improve MDT with a more efficient macro network structure and training strategy, named MDTv2. Experimental results show that MDTv2 achieves superior image synthesis performance, e.g., a new SOTA FID score of 1.58 on the ImageNet dataset, and has more than 10x faster learning speed than the previous SOTA DiT. The source code is released at https://github.com/sail-sg/MDT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2303_14389 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | MDTv2: Masked Diffusion Transformer is a Strong Image Synthesizer Gao, Shanghua Zhou, Pan Cheng, Ming-Ming Yan, Shuicheng Computer Vision and Pattern Recognition Despite its success in image synthesis, we observe that diffusion probabilistic models (DPMs) often lack contextual reasoning ability to learn the relations among object parts in an image, leading to a slow learning process. To solve this issue, we propose a Masked Diffusion Transformer (MDT) that introduces a mask latent modeling scheme to explicitly enhance the DPMs' ability to contextual relation learning among object semantic parts in an image. During training, MDT operates in the latent space to mask certain tokens. Then, an asymmetric diffusion transformer is designed to predict masked tokens from unmasked ones while maintaining the diffusion generation process. Our MDT can reconstruct the full information of an image from its incomplete contextual input, thus enabling it to learn the associated relations among image tokens. We further improve MDT with a more efficient macro network structure and training strategy, named MDTv2. Experimental results show that MDTv2 achieves superior image synthesis performance, e.g., a new SOTA FID score of 1.58 on the ImageNet dataset, and has more than 10x faster learning speed than the previous SOTA DiT. The source code is released at https://github.com/sail-sg/MDT. |
| title | MDTv2: Masked Diffusion Transformer is a Strong Image Synthesizer |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2303.14389 |