@@ -32,14 +32,31 @@ results pave the way for generative models better suited for video and animation
3232 <img src =" https://user-images.githubusercontent.com/22982797/150450502-c182834f-796f-4397-bd38-df1efe4a8a47.png " width =" 800 " />
3333</div >
3434
35+ We perform experiments on StyleGANv3 paper settings and also experimental settings.
36+ For user convenience, we also offer the converted version of official weights.
37+
38+ ### Paper Settings
39+ | Model | Dataset | Iter | FID50k | Config | Log | Download |
40+ | :---------------------------------: | :-------------: | :-----------: | :-----------: | :---------------------------------------------------------------------------------------------------------------------------: | :-------------: | :--------------------------------------------------------------------------------------------------------------------------------------: |
41+ | stylegan3-t | ffhq 1024x1024 | 490000 | 3.37<sup >* </sup >| [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/styleganv3/stylegan3_t_noaug_fp16_gamma32.8_ffhq_1024_b4x8.py ) | [ log] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_noaug_fp16_gamma32.8_ffhq_1024_b4x8_20220322_090417.log.json ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_noaug_fp16_gamma32.8_ffhq_1024_b4x8_best_fid_iter_490000_20220401_120733-4ff83434.pth ) |
42+ | stylegan3-t-ada | metface 1024x1024 | 130000 | 15.09 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/styleganv3/stylegan3_t_ada_fp16_gamma6.6_metfaces_1024_b4x8.py ) | [ log] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_ada_fp16_gamma6.6_metfaces_1024_b4x8_20220328_142211.log.json ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_ada_fp16_gamma6.6_metfaces_1024_b4x8_best_fid_iter_130000_20220401_115101-f2ef498e.pth ) |
43+
44+ Note<sup >* </sup >: This setting still needs a few days to run through, we put out currently the best checkpoint, and we will update the results the first time on the end of the experiment.
45+
46+ ### Experimental Settings
47+ | Model | Dataset | Iter | FID50k | Config | Log | Download |
48+ | :---------------------------------: | :-------------: | :-----------: | :-----------: | :---------------------------------------------------------------------------------------------------------------------------: | :-------------: | :--------------------------------------------------------------------------------------------------------------------------------------: |
49+ | stylegan3-t | ffhq 256x256 | 740000 | 7.65 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/styleganv3/stylegan3_t_noaug_fp16_gamma2.0_ffhq_256_b4x8.py ) | [ log] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_noaug_fp16_gamma2.0_ffhq_256_b4x8_20220323_144815.log.json ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_noaug_fp16_gamma2.0_ffhq_256_b4x8_best_fid_iter_740000_20220401_122456-730e1fba.pth ) |
50+
51+ ### Converted Weights
3552| Model | Dataset | Comment | FID50k | EQ-T | EQ-R | Config | Download |
3653| :---------------------------------: | :-------------: | :-------------: | :----: | :-----------: | :-----------: | :---------------------------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------: |
37- | stylegan3_config-T | ffhqu 256x256| official weight | 4.62 | 63.01 | 13.12 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_t_ffhqu_256_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_ffhqu_256_b4x8_cvt_official_rgb_20220329_235046-153df4c8.pth ) |
38- | stylegan3_config-T | afhqv2 512x512 | official weight | 4.04 | 60.15 | 13.51 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_t_afhqv2_512_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_afhqv2_512_b4x8_cvt_official_rgb_20220329_235017-ee6b037a.pth ) |
39- | stylegan3_config-T | ffhq 1024x1024 | official weight | 2.79 | 61.21 | 13.82 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_t_ffhq_1024_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_ffhq_1024_b4x8_cvt_official_rgb_20220329_235113-db6c6580.pth ) |
40- | stylegan3_config-R | ffhqu 256x256 | official weight | 4.50| 66.65 | 40.48 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_r_ffhqu_256_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_r_ffhqu_256_b4x8_cvt_official_rgb_20220329_234909-4521d963.pth ) |
41- | stylegan3_config-R | afhqv2 512x512 | official weight | 4.40 | 64.89 | 40.34 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_r_afhqv2_512_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_r_afhqv2_512_b4x8_cvt_official_rgb_20220329_234829-f2eaca72.pth ) |
42- | stylegan3_config-R | ffhq 1024x1024 | official weight | 3.07 | 64.76 | 46.62 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_r_ffhq_1024_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_r_ffhq_1024_b4x8_cvt_official_rgb_20220329_234933-ac0500a1.pth ) |
54+ | stylegan3-t | ffhqu 256x256| official weight | 4.62 | 63.01 | 13.12 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_t_ffhqu_256_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_ffhqu_256_b4x8_cvt_official_rgb_20220329_235046-153df4c8.pth ) |
55+ | stylegan3-t | afhqv2 512x512 | official weight | 4.04 | 60.15 | 13.51 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_t_afhqv2_512_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_afhqv2_512_b4x8_cvt_official_rgb_20220329_235017-ee6b037a.pth ) |
56+ | stylegan3-t | ffhq 1024x1024 | official weight | 2.79 | 61.21 | 13.82 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_t_ffhq_1024_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_t_ffhq_1024_b4x8_cvt_official_rgb_20220329_235113-db6c6580.pth ) |
57+ | stylegan3-r | ffhqu 256x256 | official weight | 4.50| 66.65 | 40.48 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_r_ffhqu_256_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_r_ffhqu_256_b4x8_cvt_official_rgb_20220329_234909-4521d963.pth ) |
58+ | stylegan3-r | afhqv2 512x512 | official weight | 4.40 | 64.89 | 40.34 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_r_afhqv2_512_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_r_afhqv2_512_b4x8_cvt_official_rgb_20220329_234829-f2eaca72.pth ) |
59+ | stylegan3-r | ffhq 1024x1024 | official weight | 3.07 | 64.76 | 46.62 | [ config] ( https://github.com/open-mmlab/mmgeneration/tree/master/configs/_base_/models/stylegan/stylegan3_r_ffhq_1024_b4x8_cvt_official_rgb.py ) | [ model] ( https://download.openmmlab.com/mmgen/stylegan3/stylegan3_r_ffhq_1024_b4x8_cvt_official_rgb_20220329_234933-ac0500a1.pth ) |
4360
4461
4562
@@ -73,7 +90,7 @@ https://user-images.githubusercontent.com/22982797/151504973-b96e1639-861d-434b-
7390
7491https://user-images.githubusercontent.com/22982797/151505099-cde4999e-aab1-42d4-a458-3bb069db3d32.mp4
7592
76- If you want to EQ-Metric for StyleGAN3, just add following codes into config.
93+ If you want to get EQ-Metric for StyleGAN3, just add following codes into config.
7794``` python
7895metrics = dict (
7996 eqv = dict (
0 commit comments