3gi's repositories
rk3588-npu
Reverse engineering the rk3588 npu
MoeVoiceStudio
一个使用C++编写的音频处理软件
Bark-Voice-Cloning
Bark Voice Cloning and Voice Cloning for Chinese Speech
FaceAlgorithm
face detection face recognition包含人脸检测(retinaface,yolov5face,yolov7face,yolov8face),人脸检测跟踪(ByteTracker)，人脸角度计算（Face_Angle）人脸矫正(Face_Aligner)，人脸识别(Arcface)，口罩检测(MaskRecognitiion)，年龄性别检测(Gender_age)，静默活体检测(Silent_Face_Anti_Spoofing),FaceAlignment(106keypoints)
PlateRecognition
License-Plate-Recognition 支持12种车牌检测识别，包含yolov5,yolov7,yolov8车牌检测，车牌矫正，车牌识别等，准确率高达99.5% 还有车牌数据集提供下载
CLIP-ONNX-AX650-CPP
c++实现的clip推理，模型有一点点改动，但是不大，改动和导出模型的代码可以在readme里找到，模型文件都在Releases里，包括AX650的模型。新增支持ChineseCLIP
sherpa-ncnn
Real-time speech recognition using next-gen Kaldi with ncnn without Internet connection. Support iOS, Android, Raspberry Pi, VisionFive2, etc.
multitask-hydranet
multitask mono camera perception with three tasks: object detection; lane detection; semantic segmentation
onnx2pytorch
Transform ONNX model to PyTorch representation
MVision
机器人视觉 移动机器人 VS-SLAM ORB-SLAM2 深度学习目标检测 yolov3 行为检测 opencv PCL 机器学习 无人驾驶
model-compression
model compression based on pytorch (1、quantization: 8/4/2bits(dorefa)、ternary/binary value(twn/bnn/xnor-net)；2、 pruning: normal、regular and group convolutional channel pruning；3、 group convolution structure；4、batch-normalization folding for binary value of feature(A))
models
Models and examples built with TensorFlow
BinaryNet
Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
mediasrv_linux
流媒体服务器Linux版本 RTMP HLS RTSP Stream server。 Support HLS RTMP RTSP HTTP-TS HTTP-FLV HTTP-AAC。 Output video in 0.2-0.3 seconds，very fast。