The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more
- pytorch
- resnet
- pretrained-models
- pretrained-weights
TranslatedUltra-lightweight Chinese OCR, supports vertical text recognition, supports ncnn, mnn, and tnn inference (dbnet (1.8M) + crnn (2.5M) + anglenet (378KB)), total model size only 4.7M
Implementation of popular deep learning networks with TensorRT network definition API
- tensorrt
- mnasnet
- resnet
- squeezenet
NanoDet-Plus⚡Super fast and lightweight anchor-free object detection model. 🔥Only 980 KB(int8) / 1.8MB (fp16) and run 97FPS on cellphone🔥
Segmentation models with pretrained backbones. Keras and TensorFlow Keras.
- unet
- fpn
- segmentation
- keras
Practice on cifar100(ResNet, DenseNet, VGG, GoogleNet, InceptionV3, InceptionV4, Inception-ResNetv2, Xception, Resnet In Resnet, ResNext,ShuffleNet, ShuffleNetv2, MobileNet, MobileNetv2, SqueezeNet, NasNet, Residual Attention Network, SENet, WideResNet)
OpenMMLab Pre-training Toolbox and Benchmark
Tensorflow Faster RCNN for Object Detection
DeepLab v3+ model in PyTorch. Support different backbones.
- deeplab-v3-plus
- pytorch
- xception
- resnet
Retinaface get 80.99% in widerface hard val using mobilenet0.25.
🍅🍅🍅YOLOv5-Lite: Evolved from yolov5 and the size of model is only 900+kb (int8) and 1.7M (fp16). Reach 15 FPS on the Raspberry Pi 4B~
Multiple Object Tracker, Based on Hungarian algorithm + Kalman filter.
- multiple-object-tracking
- kalman-filter
- hungarian-algorithm
- car-tracking
YOLO5Face: Why Reinventing a Face Detector (https://arxiv.org/abs/2105.12931) ECCV Workshops 2022)
- yolov5
- tensorrt
- retinaface
- scrfd
A Super-fast account switcher for Steam, Battle.net, Epic Games, Origin, Riot, Ubisoft and many others!
- tcno-account-switcher
- account-switcher
- steam
- origin
Caffe implementation of Google MobileNet SSD detection network, with pretrained weights on VOC0712 and mAP=0.727.
- ssd
- mobilenet-ssd
- caffe
- mobilenet
Robust realtime face and facial landmark tracking on CPU with Unity integration
- face-tracking
- face-landmarks
- depth-estimation
- unity
Implementation of EfficientNet model. Keras and TensorFlow Keras.
- classification
- imagenet
- efficientnet
- pretrained-models
mobilenetv3 with pytorch,provide pre-train model
中文长文本分类、短句子分类、多标签分类、两句子相似度(Chinese Text Classification of Keras NLP, multi-label classify, or sentence classify, long or short),字词句向量嵌入层(embeddings)和网络层(graph)构建基类,FastText,TextCNN,CharCNN,TextRNN, RCNN, DCNN, DPCNN, VDCNN, CRNN, Bert, Xlnet, Albert, Attention, DeepMoji, HAN, 胶囊网络-CapsuleNet, Transformer-encode, Seq2seq, SWEM, LEAM, TextGCN
- text-classification
- keras
- rcnn
- dcnn
TensorRT MODNet, YOLOv4, YOLOv3, SSD, MTCNN, and GoogLeNet
- tensorrt
- yolov4
- yolov3
- ssd-mobilenet
MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire:
- cnn
- yolov3
- yolo
- mobilenetv2
Easy training on custom dataset. Various backends (MobileNet and SqueezeNet) supported. A YOLO demo to detect raccoon run entirely in brower is accessible at https://git.io/vF7vI (not on Windows).
This is a pytorch repository of YOLOv4, attentive YOLOv4 and mobilenet YOLOv4 with PASCAL VOC and COCO
MobileNet build with Tensorflow