search

AlvinYH / Faster-VoxelPose

Official implementation of Faster VoxelPose: Real-time 3D Human Pose Estimation by Orthographic Projection
MIT License
154 stars 18 forks source link

Anyone can tell me about train time? #28

Open qhdqhd opened 1 year ago

qhdqhd commented 1 year ago

How long does it need for training? Thanks!

AlvinYH commented 1 year ago

Thanks for your interest in our work. We conducted our experiments on a single NVIDIA GeForce RTX 2080 GPU. For the Panoptic dataset, it takes ~24 hours for training 10 epochs. As for the Campus/Shelf dataset, we synthesize 10K training samples, and it takes ~30min for training 1 epoch (usually take 20-30 epochs for convergence).