TTNet: Real-time temporal and spatial video analysis of table tennis
TTNet: Real-time temporal and spatial video analysis of table tennis
I implemented the paper that published at CVPR 2020 from scratch.
Source code: https://github.com/maudzung/TTNet-Realtime-for-Table-Tennis-Pytorch
Source paper: https://arxiv.org/pdf/2004.09927.pdf
Source video is here
Ball detection global stage
Ball detection local stage (refinement)
Events Spotting detection (Bounce and Net hit)
Semantic Segmentation (Human, table, and scoreboard)
Enable/Disable modules in the TTNet model
Accuracy for the event spotting task
IoU for the segmentation task
RMSE for the ball detection task