MobileI2V: Fast and High-Resolution Image-to-Video on Mobile Devices
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_ | 1866917106613747712 |
|---|---|
| author | Zhang, Shuai Tang, Bao Yu, Siyuan Zhu, Yueting Yao, Jingfeng Zou, Ya Yuan, Shanglin Yu, Li Liu, Wenyu Wang, Xinggang |
| author_facet | Zhang, Shuai Tang, Bao Yu, Siyuan Zhu, Yueting Yao, Jingfeng Zou, Ya Yuan, Shanglin Yu, Li Liu, Wenyu Wang, Xinggang |
| contents | Recently, video generation has witnessed rapid advancements, drawing increasing attention to image-to-video (I2V) synthesis on mobile devices. However, the substantial computational complexity and slow generation speed of diffusion models pose significant challenges for real-time, high-resolution video generation on resource-constrained mobile devices. In this work, we propose MobileI2V, a 270M lightweight diffusion model for real-time image-to-video generation on mobile devices. The core lies in: (1) We analyzed the performance of linear attention modules and softmax attention modules on mobile devices, and proposed a linear hybrid architecture denoiser that balances generation efficiency and quality. (2) We design a time-step distillation strategy that compresses the I2V sampling steps from more than 20 to only two without significant quality loss, resulting in a 10-fold increase in generation speed. (3) We apply mobile-specific attention optimizations that yield a 2-fold speed-up for attention operations during on-device inference. MobileI2V enables, for the first time, fast 720p image-to-video generation on mobile devices, with quality comparable to existing models. Under one-step conditions, the generation speed of each frame of 720p video is less than 100 ms. Our code is available at: https://github.com/hustvl/MobileI2V. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_21475 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MobileI2V: Fast and High-Resolution Image-to-Video on Mobile Devices Zhang, Shuai Tang, Bao Yu, Siyuan Zhu, Yueting Yao, Jingfeng Zou, Ya Yuan, Shanglin Yu, Li Liu, Wenyu Wang, Xinggang Computer Vision and Pattern Recognition Recently, video generation has witnessed rapid advancements, drawing increasing attention to image-to-video (I2V) synthesis on mobile devices. However, the substantial computational complexity and slow generation speed of diffusion models pose significant challenges for real-time, high-resolution video generation on resource-constrained mobile devices. In this work, we propose MobileI2V, a 270M lightweight diffusion model for real-time image-to-video generation on mobile devices. The core lies in: (1) We analyzed the performance of linear attention modules and softmax attention modules on mobile devices, and proposed a linear hybrid architecture denoiser that balances generation efficiency and quality. (2) We design a time-step distillation strategy that compresses the I2V sampling steps from more than 20 to only two without significant quality loss, resulting in a 10-fold increase in generation speed. (3) We apply mobile-specific attention optimizations that yield a 2-fold speed-up for attention operations during on-device inference. MobileI2V enables, for the first time, fast 720p image-to-video generation on mobile devices, with quality comparable to existing models. Under one-step conditions, the generation speed of each frame of 720p video is less than 100 ms. Our code is available at: https://github.com/hustvl/MobileI2V. |
| title | MobileI2V: Fast and High-Resolution Image-to-Video on Mobile Devices |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.21475 |