|
pulsatrix
|
Samples num_frames evenly-spaced frames from a clip Tensor (sample.fields[0], shape (T, C, H, W) – VideoFrameDirectoryDataset's convention), producing an (N, C, H, W) Tensor – deliberately identical in shape convention to Phase 2's image Dataset/Transform outputs (ImageFolderDataset, ResizeTransform, etc.): a video clip, once frame-sampled, is just a batch of images from this codebase's perspective. More...
#include <video_transforms.hpp>


Public Member Functions | |
| UniformFrameSampleTransform (int64_t num_frames, DeviceBackend *backend) | |
| Sample | apply (Sample sample) const override |
Public Member Functions inherited from pulsatrix::Transform | |
| virtual | ~Transform ()=default |
Samples num_frames evenly-spaced frames from a clip Tensor (sample.fields[0], shape (T, C, H, W) – VideoFrameDirectoryDataset's convention), producing an (N, C, H, W) Tensor – deliberately identical in shape convention to Phase 2's image Dataset/Transform outputs (ImageFolderDataset, ResizeTransform, etc.): a video clip, once frame-sampled, is just a batch of images from this codebase's perspective.
|
inline |
| std::invalid_argument | if num_frames exceeds the clip's actual frame count – external boundary: num_frames is caller-supplied configuration. |
Implements pulsatrix::Transform.