Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.06430 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908482524938240 |
|---|---|
| author | Patil, Om Modi, Jinesh Mukhopadhyay, Suryabha Giri, Meghaditya Malhotra, Chhavi |
| author_facet | Patil, Om Modi, Jinesh Mukhopadhyay, Suryabha Giri, Meghaditya Malhotra, Chhavi |
| contents | Face swapping technology has gained significant attention in both academic research and commercial applications. This paper presents our implementation and enhancement of SimSwap, an efficient framework for high fidelity face swapping. We introduce several improvements to the original model, including the integration of self and cross-attention mechanisms in the generator architecture, dynamic loss weighting, and cosine annealing learning rate scheduling. These enhancements lead to significant improvements in identity preservation, attribute consistency, and overall visual quality.
Our experimental results, spanning 400,000 training iterations, demonstrate progressive improvements in generator and discriminator performance. The enhanced model achieves better identity similarity, lower FID scores, and visibly superior qualitative results compared to the baseline. Ablation studies confirm the importance of each architectural and training improvement. We conclude by identifying key future directions, such as integrating StyleGAN3, improving lip synchronization, incorporating 3D facial modeling, and introducing temporal consistency for video-based applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_06430 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MotionSwap Patil, Om Modi, Jinesh Mukhopadhyay, Suryabha Giri, Meghaditya Malhotra, Chhavi Computer Vision and Pattern Recognition Face swapping technology has gained significant attention in both academic research and commercial applications. This paper presents our implementation and enhancement of SimSwap, an efficient framework for high fidelity face swapping. We introduce several improvements to the original model, including the integration of self and cross-attention mechanisms in the generator architecture, dynamic loss weighting, and cosine annealing learning rate scheduling. These enhancements lead to significant improvements in identity preservation, attribute consistency, and overall visual quality. Our experimental results, spanning 400,000 training iterations, demonstrate progressive improvements in generator and discriminator performance. The enhanced model achieves better identity similarity, lower FID scores, and visibly superior qualitative results compared to the baseline. Ablation studies confirm the importance of each architectural and training improvement. We conclude by identifying key future directions, such as integrating StyleGAN3, improving lip synchronization, incorporating 3D facial modeling, and introducing temporal consistency for video-based applications. |
| title | MotionSwap |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.06430 |