3 results found Sort:

v-iashin
video_features v-iashin
97
552
mit
6
Extract video features from raw videos using multiple GPUs. We support RAFT flow frames as well as S3D, I3D, R(2+1)D, VGGish, CLIP, and TIMM models.
i3d s3d vit clip raft swin timm ig65m laion resnet vggish pytorch parallel r2plus1d multi-gpu optical-flow audio-features video-features visual-features feature-extraction
Created 2020-06-01
98 commits to master branch, last one about a month ago
naver
kapture naver
66
481
bsd-3-clause
21
kapture is a file format as well as a set of tools for manipulating datasets, and in particular Visual Localization and Structure from Motion data.
sfm r2d2 colmap localization image-retrieval visual-features visual-localization structure-from-motion
Created 2020-06-26
896 commits to main branch, last one 9 months ago
multimodal
multimodal multimodal
7
79
mit
4
A collection of multimodal datasets, and visual features for VQA and captionning in pytorch. Just run "pip install multimodal"
datasets reasoning embeddings deep-learning visual-features vision-and-language
Created 2020-03-13
143 commits to master branch, last one 2 years ago