Perfect implement. Model shared. x0.5 (Top1:60.646) and 1.0x (Top1:69.402).
The main features of ericsun99/shufflenet-v2-pytorch are: Computer Vision Models, Facial, Action and Pose Recognition.
Open-source alternatives to ericsun99/shufflenet-v2-pytorch include: nvidia/flownet2-pytorch — Pytorch implementation of FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks. kenshohara/3d-resnets-pytorch — This project is a PyTorch implementation of 3D residual networks designed for video action recognition. It provides a… 1adrianb/face-alignment — This is a PyTorch-based computer vision library for detecting 2D and 3D facial landmark coordinates. It functions as a… davexpro/pytorch-pose-estimation — PyTorch Implementation of Realtime Multi-Person Pose Estimation project. aaltovision/dgc-net — A PyTorch implementation of "DGC-Net: Dense Geometric Correspondence Network". thnkim/openfacepytorch — PyTorch module to use OpenFace's nn4.small2.v1.t7 model.
PyTorch Implementation of Realtime Multi-Person Pose Estimation project.
A PyTorch implementation of "DGC-Net: Dense Geometric Correspondence Network"
This is a PyTorch-based computer vision library for detecting 2D and 3D facial landmark coordinates. It functions as a facial landmark detector and reconstruction tool, utilizing deep learning to identify precise geometric points on human faces from image datasets. The library allows for the selection of specific detection backends to balance accuracy and processing speed. It supports the integration of precomputed bounding box files, which enables the system to bypass the initial detection phase and proceed directly to landmark extraction. The toolkit includes capabilities for batch image p
This project is a PyTorch implementation of 3D residual networks designed for video action recognition. It provides a spatiotemporal architecture that analyzes both spatial frames and temporal motion to classify human activities within video clips. The system includes a distributed model training framework to accelerate learning across multiple compute nodes. It supports the deployment and fine-tuning of pre-trained model weights, allowing the adaptation of existing networks to specific new datasets. The codebase covers the full pipeline for spatiotemporal learning, including video dataset p