qaz734913414 / MNN_FaceTrackLinks
开源视频人脸跟踪算法,MNN基于mtcnn人脸检测+onet人脸跟踪,在i7-9700k的cpu检测速度可高达500fps
☆23Updated 5 years ago
Alternatives and similar repositories for MNN_FaceTrack
Users that are interested in MNN_FaceTrack are comparing it to the libraries listed below
Sorting:
- 移动端快速人脸检测模型是基于RetinaFace的优化去掉stride8以及stride32和stride16的landmark 在CPU位Intel(R) Pentium(R) CPU G2020 @ 2.90GHz(2900 MHz)的设备 中人脸检测可达到40ms/帧☆15Updated 6 years ago
- A re-implementation of PFLD, https://arxiv.org/abs/1902.10859☆45Updated 6 years ago
- train Gender and Age☆39Updated 7 years ago
- ☆16Updated 5 years ago
- ☆30Updated 6 years ago
- ☆10Updated 5 years ago
- Based on Onet to stabilize Face Detection BoundingBox. Fast, Smooth.☆61Updated 3 years ago
- MobileNet-SSD Face Detection by ncnn☆21Updated 7 years ago
- fast face detector☆18Updated 6 years ago
- pytorch face_landmark☆25Updated 2 years ago
- 「Pytorch」<PFLD: A Practical Facial Landmark Detector>☆72Updated 3 years ago
- train ssd☆10Updated 6 years ago
- 用pytorch训练ssd,相比原版pytorch-ssd改动了不少☆11Updated 3 years ago
- Real-time iris detector. Only need 8ms on Intel i5 CPU!☆21Updated 6 years ago
- mnn based mtcnn c++ realize.☆30Updated 6 years ago
- caffe train face licenseplate reID action ocr centernet☆23Updated 5 years ago
- Retinaface caffe with mnetv2 0.25(and many more)☆57Updated 6 years ago
- ☆34Updated 6 years ago
- train mxnet unet, then run it in ncnn☆64Updated 7 years ago
- A C++ API of the LFFD with ncnn☆99Updated 5 years ago
- ☆11Updated 6 years ago
- efficient header for third-party libs☆12Updated 3 years ago
- ☆16Updated 5 years ago
- MNN MTCNN C++☆50Updated 6 years ago
- SuperResolution\SR\GPU\Traditional image processing☆20Updated 7 years ago
- faceboxes@ncnn☆41Updated 6 years ago
- reproduce faceboxes☆53Updated 6 years ago
- MTCNN light + SORT tracking☆44Updated 5 years ago
- deep-sdm is appied for face landmark.☆73Updated 5 years ago
- Lightweight face detectors with landmarks. Training code using pytorch and inference using pytorch/ncnn/tensorflow/tflite.☆10Updated 5 years ago