kiui

@tylersky1993 Hi, if you are going to process the data by yourself, you need to follow the steps [here](https://github.com/ashawkey/RAD-NeRF#preparation) first.

@Souvic Hi, what is the dataset? If the head still shows severe flickering at a fixed pose, you can try to increase [`lambda_amb`](https://github.com/ashawkey/RAD-NeRF/blob/main/main.py#L38) in training, which regularize unnecessary dynamics outside...

@WuNvRenQiaoDaiMaZiRanShen Hi, currently we just repeat the head poses in training sequence. It's better to predict a head pose sequence based on audio, but this is not done in this...

@snicolast Hi, I have obeserved this too in the second stage. I still haven't figured out why, but you may try to decrease `iters_refine` to alleviate it: https://github.com/dreamgaussian/dreamgaussian/blob/main/configs/text_mv.yaml#L30C6-L30C6

These tractors seem good to me? You may check the first stage model too. If it is already green, then the second stage may not change the color. That error...

@hodakagoto Hi, yes! You can use the `load` config to directly load the `ply` files.

@ksupasate Hi, you should install it first, please follow the readme and make sure each step in `install` succeed.

@Fanie-Visagie Hi, please update `diffusers` and `transformers` to the latest.

@nagolinc Hi, please check https://github.com/dreamgaussian/dreamgaussian/issues/56, zero123 plus only generates fixed 6 views and cannot be applied here.

@maybeLx Hi, actually I'm confused with gaussian splattings' coordinate system too, so I just align the coordinate system in a brute-force way.