Train action classification model based on individual frames
仓库
gurkirt 的仓库
3D-RetinaNet a baseline models on ROAD dataset
Action Micro Tube Network (AMTNet) - Pytorch with linear heads
Feature pyramid network (FPN) with online hard example mining (OHEM)
RetinaNet with different loss function types
PySlowFast: video understanding codebase from FAIR for reproducing state-of-the-art video models.
This repository is intended to host tools and demos for ActivityNet
Action tracker from GTR
This repo provided utility tools to browse the videos and annotation files for DALY dataset.
Tensor-flow trainer for action classification and detection
Differentiable-winograd implementation in PyTorch
Diffuse optical flow for 3D edits of images
Download kinetics and prepare '.json' annotations for each subset
Official Implementation of CVPR24 highligt paper: Matching Anything by Segmenting Anything
compute optical flow on GPU using opencv and set of image sequences rather than videos. Use ffmpeg for video to image sequence conversion