search

karfly / learnable-triangulation-pytorch

This repository is an official PyTorch implementation of the paper "Learnable Triangulation of Human Pose" (ICCV 2019, oral). Proposed method archives state-of-the-art results in multi-view 3D human pose estimation!
MIT License
1.1k stars 181 forks source link

Question: Will pre processing all frames (instead of skipping frames) make it temporally smoother? ie. less jittering #88

Closed jamalknight closed 4 years ago

jamalknight commented 4 years ago

The pre-processing script skips frames and does not process every frame. Would this affect the temporal movement of the moving pose over time? ie. If all frames were processed, would the resulting animation be smoother, or would this not be an issue at this stage? -perhaps a temporal smoothing is done in steps further down? Just curious

shrubb commented 4 years ago

Our method doesn't do any temporal smoothing, i.e. we process a frame independently of others. So no, skipping is all right, it's done to remove too similar frames that are redundant for learning.

jamalknight commented 4 years ago

Thanks for clarifying.