XiaPengcheng

Is there any method to change BFM without neck to FLAME

4

Since I use the [3DDFA_V2](https://github.com/cleardusk/3DDFA_V2) to predict BFM model without neck area, how can i change it to FLAME model. BFM model without neck part only has 35709 points.

│RuntimeError: CUDA error: out of memory

1

I try to reproduce the result with 4 1080TI ,and i reduce the batch size=4, The detailed configuration `python main.py \ --exp-name line_weighted_wo_focal_junc --backbone resnet50 \ --backbone-kwargs '{"encoder_weights": "ckpt/backbone/encoder_epoch_20.pth", "decoder_weights":...

Train rq-vae on FFHQ failed when using the default ffhq256-rqvae-8x8x4.yaml

2

I trt to retrain rq-vae on FFHQ with the default ffhq256-rqvae-8x8x4.yaml. The training loss first decrease and then increase.  and Then I compute rfid using the model with the...

Can I use the code to train talking head generation model just like diffuse heads!

1

Since the training pipeline of diffuse head is unclear, can I use the code to train talking head generation model just like diffuse heads!