Context Encoder

[fsnelgar]

In the paper a 'Context Encoder' is used to condition the diffusion model.
It appears that this conditioning is here https://github.com/GONGJIA0208/Diffpose/blob/main/models/gcndiff.py#L101 but doesn't appear to be used. Is this the case?

[GoogQing]

Yes, we just use 'u v' as the condition in the frame-based experiment, which is strong enough.
The feature from the context encoder is mainly used for video-based experiments which require dense spatial-temporal information.

When we have time in recent months, we will update the model for the video-based pipeline. Thanks :)

[gtftyih]

是的，我们只是在基于框架的实验中使用“u v”作为条件，这已经足够强大了。上下文编码器的功能主要用于需要密集时空信息的基于视频的实验。

最近几个月有时间时，我们将更新基于视频的管道的模型。谢谢:)

Yes, we just use 'u v' as the condition in the frame-based experiment, which is strong enough. The feature from the context encoder is mainly used for video-based experiments which require dense spatial-temporal information.

When we have time in recent months, we will update the model for the video-based pipeline. Thanks :)

Hello，does ‘GCNpose’ just get the 3D pose directly?