C3D
C3d is a modified version of BVLC caffe to support 3D convolution and pooling.
The main supporting features include:
- Training or fine-tuning 3D ConvNets.
- Extracting video features with pre-trained C3D models.
For more information about c3d, please refer to the C3D project website.
For general questions about Caffe, please refer to the BVLC project website for all documentation.
Third-party Implementation:
- C3D-tensorflow with trained models by hx and XL