fwbx529

> > 阿里的开源项目，代码一般会延迟两个月的时间发布，内部走流程啥的。 当然不能完全否定会放鸽子。 > > 这个阿里另外一个团队开源项目dreaMoving的首次code提交日期：Commits on Dec 7, 2023 dreaMoving在12月31号已经开放了demo试用链接：https://www.modelscope.cn/studios/vigen/video_generation/summary 从首次提交到开放demo用了不到一个月时间，反观animate-anyone两个月了，除了star数，啥进展都没有 今天听说有demo了，通义千问app输入全民舞王可以试用

I use groundsam/{}.mask.png. The generated png is not pure black, the fg area has the RGB value of (1, 1, 1)

Hi, I have a similar problem. I run the evaluation using this script (with [More TikTok-Style Training Data (FID-FVD: 15.7)](https://storage.googleapis.com/disco-checkpoint-share/checkpoint_ft/moretiktok_cfg/mp_rank_00_model_states.pt)): `AZFUSE_USE_FUSE=0 NCCL_ASYNC_ERROR_HANDLING=0 python finetune_sdm_yaml.py --cf config/ref_attn_clip_combine_controlnet/tiktok_S256L16_xformers_tsv.py --eval_visu --root_dir run_test --local_train_batch_size...

BTW, I found that the gif generation using imageio at 3fps (gen_eval.sh) will cause tbr as 24.25, so when ffmpeg change it to video, the origin 16 frame gif will...

Hi, sorry for the late reply, as I am currently working on another project here, not focusing on video gen. When I tested FVD before, I used an online-downloaded ckpt...