CondiQuant: Condition Number Based Low-Bit Quantization for Image Super-Resolution
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_ | 1866929724816621568 |
|---|---|
| author | Liu, Kai Wang, Dehui Li, Zhiteng Chen, Zheng Guo, Yong Li, Wenbo Kong, Linghe Zhang, Yulun |
| author_facet | Liu, Kai Wang, Dehui Li, Zhiteng Chen, Zheng Guo, Yong Li, Wenbo Kong, Linghe Zhang, Yulun |
| contents | Low-bit model quantization for image super-resolution (SR) is a longstanding task that is renowned for its surprising compression and acceleration ability. However, accuracy degradation is inevitable when compressing the full-precision (FP) model to ultra-low bit widths (2~4 bits). Experimentally, we observe that the degradation of quantization is mainly attributed to the quantization of activation instead of model weights. In numerical analysis, the condition number of weights could measure how much the output value can change for a small change in the input argument, inherently reflecting the quantization error. Therefore, we propose CondiQuant, a condition number based low-bit post-training quantization for image super-resolution. Specifically, we formulate the quantization error as the condition number of weight metrics. By decoupling the representation ability and the quantization sensitivity, we design an efficient proximal gradient descent algorithm to iteratively minimize the condition number and maintain the output still. With comprehensive experiments, we demonstrate that CondiQuant outperforms existing state-of-the-art post-training quantization methods in accuracy without computation overhead and gains the theoretically optimal compression ratio in model parameters. Our code and model are released at https://github.com/Kai-Liu001/CondiQuant. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_15478 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CondiQuant: Condition Number Based Low-Bit Quantization for Image Super-Resolution Liu, Kai Wang, Dehui Li, Zhiteng Chen, Zheng Guo, Yong Li, Wenbo Kong, Linghe Zhang, Yulun Computer Vision and Pattern Recognition Low-bit model quantization for image super-resolution (SR) is a longstanding task that is renowned for its surprising compression and acceleration ability. However, accuracy degradation is inevitable when compressing the full-precision (FP) model to ultra-low bit widths (2~4 bits). Experimentally, we observe that the degradation of quantization is mainly attributed to the quantization of activation instead of model weights. In numerical analysis, the condition number of weights could measure how much the output value can change for a small change in the input argument, inherently reflecting the quantization error. Therefore, we propose CondiQuant, a condition number based low-bit post-training quantization for image super-resolution. Specifically, we formulate the quantization error as the condition number of weight metrics. By decoupling the representation ability and the quantization sensitivity, we design an efficient proximal gradient descent algorithm to iteratively minimize the condition number and maintain the output still. With comprehensive experiments, we demonstrate that CondiQuant outperforms existing state-of-the-art post-training quantization methods in accuracy without computation overhead and gains the theoretically optimal compression ratio in model parameters. Our code and model are released at https://github.com/Kai-Liu001/CondiQuant. |
| title | CondiQuant: Condition Number Based Low-Bit Quantization for Image Super-Resolution |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.15478 |