search

eververas / 3DGazeNet

[ECCV 2024] 3DGazeNet: Generalizing Gaze Estimation with Weak-Supervision from Synthetic Views
https://arxiv.org/abs/2212.02997
57 stars 5 forks source link

Input video? #5

Open Echo-jyt opened 6 months ago

Echo-jyt commented 6 months ago

Can video be used as input ？

eververas commented 5 months ago

Hello, you can use the demo code from here https://github.com/Vagver/3DGazeNet/tree/main/demo to directly get output from videos. Alternatively, you can split your video into frames and use inference.py from the main directory.