Irene

Benchmark_model provided seems ineffective on gpu

3

Hi, when I profile the gpu latency on Snapdragon888+ with tf2.7 benchmark_model you provided, the latency seems to be always zero. Is there any idea? Thank you in advance!

Weight of Poseguider changed extremely small

1

I set lr=1e-5, the weight of poseguider changed little comparing with initial, is this correct? or we just need to waiting for more steps?

Train on Flicker8k Dataset

2

I want to train on flicker8k dataset. Some error occur in 'prepro_feats.py' when I using 'dataset_flickr8k.json'. The error is about no 'cocoid'. How to solve it? Do i need to...

inference for stage1

1

Will you public infer.py for stage1?

is any category of object support in this work?

1

Hi, fantastic work! is this work for any generic 3D object? When I input a cat image , the mesh result is just like a chair.

how to fuso the handpose paramter into SMPL-H or SMPL-X?

2

I input the rotmat directly into SMPLH handpose parameter, but it doesn't match. So how to trans the rotmat data?