All Projects → zeusees → Faceboxes

zeusees / Faceboxes

Licence: other
FaceBoxes: A CPU Real-time Face Detector with High Accuracy

Projects that are alternatives of or similar to Faceboxes

Facekit
Implementations of PCN (an accurate real-time rotation-invariant face detector) and other face-related algorithms
Stars: ✭ 1,028 (+117.34%)
Mutual labels:  face-detection, real-time
Jeelizweboji
JavaScript/WebGL real-time face tracking and expression detection library. Build your own emoticons animated in real time in the browser! SVG and THREE.js integration demos are provided.
Stars: ✭ 835 (+76.53%)
Mutual labels:  face-detection, real-time
jeelizGlanceTracker
JavaScript/WebGL lib: detect if the user is looking at the screen or not from the webcam video feed. Lightweight and robust to all lighting conditions. Great for play/pause videos if the user is looking or not, or for person detection. Link to live demo.
Stars: ✭ 68 (-85.62%)
Mutual labels:  real-time, face-detection
Realtimefaceapi
This is a demo project showing how to use Face API in Cognitive Services with OpenCV
Stars: ✭ 44 (-90.7%)
Mutual labels:  face-detection, real-time
Sod
An Embedded Computer Vision & Machine Learning Library (CPU Optimized & IoT Capable)
Stars: ✭ 1,460 (+208.67%)
Mutual labels:  face-detection, real-time
Centerface.pytorch
unofficial version of centerface, which achieves the best balance between speed and accuracy at face detection
Stars: ✭ 187 (-60.47%)
Mutual labels:  face-detection, real-time
realtime-facereccpp
Real time face recognition with tracking (mtcnn detection, kcf tracker, arcface loss)
Stars: ✭ 32 (-93.23%)
Mutual labels:  real-time, face-detection
Ksql
The database purpose-built for stream processing applications.
Stars: ✭ 4,668 (+886.89%)
Mutual labels:  real-time
Aws Mobile Appsync Chat Starter Angular
GraphQL starter progressive web application (PWA) with Realtime and Offline functionality using AWS AppSync
Stars: ✭ 449 (-5.07%)
Mutual labels:  real-time
Odas
ODAS: Open embeddeD Audition System
Stars: ✭ 435 (-8.03%)
Mutual labels:  real-time
Mobilepose Pytorch
Light-weight Single Person Pose Estimator
Stars: ✭ 427 (-9.73%)
Mutual labels:  real-time
Ray Tracing Renderer
[UNMAINTAINED] Real-time path tracing on the web with three.js
Stars: ✭ 444 (-6.13%)
Mutual labels:  real-time
Lednet
LEDNet: A Lightweight Encoder-Decoder Network for Real-time Semantic Segmentation
Stars: ✭ 450 (-4.86%)
Mutual labels:  real-time
Sofa
Real-time multi-physics simulation with an emphasis on medical simulation.
Stars: ✭ 435 (-8.03%)
Mutual labels:  real-time
Multinet
Real-time Joint Semantic Reasoning for Autonomous Driving
Stars: ✭ 471 (-0.42%)
Mutual labels:  real-time
Android Face Detector
A real-time face detection Android library
Stars: ✭ 432 (-8.67%)
Mutual labels:  face-detection
Awesome Tiny Object Detection
🕶 A curated list of Tiny Object Detection papers and related resources.
Stars: ✭ 471 (-0.42%)
Mutual labels:  face-detection
Zephyr
Primary Git Repository for the Zephyr Project. Zephyr is a new generation, scalable, optimized, secure RTOS for multiple hardware architectures.
Stars: ✭ 5,335 (+1027.91%)
Mutual labels:  real-time
Opencv4nodejs
Nodejs bindings to OpenCV 3 and OpenCV 4
Stars: ✭ 4,444 (+839.53%)
Mutual labels:  face-detection
Primus
⚡ Primus, the creator god of the transformers & an abstraction layer for real-time to prevent module lock-in.
Stars: ✭ 4,302 (+809.51%)
Mutual labels:  real-time

Faceboxes 完整复现

1. CAFFE安装:


Makefile.config已经修改了好了,使用GPU的方式
所以直接使用下面的命令编译:

    make -j8            
    # Make sure to include $CAFFE_ROOT/python to your PYTHONPATH. 
    make pycaffe        
    make test -j8       
    # (Optional)        
    make runtest -j8    

2. 数据处理:


(1) 利用脚本wider_face_2_voc.py脚本把wider_face数据转换成VOC格式。并遮盖掉小于20x20的人脸。
脚本的位置在: script

(2) 在wider_face_2_voc.py的同一级目录中创建wider_face文件夹,放解压好下载的wider数据,如图:
data

(3) 运行wider_face_2_voc.py脚本,在wider_face文件夹中会生成VOC格式的数据,如图:
data

(4) 生成的图片会把小于20x20的人脸用图像均值覆盖掉,因为太小的人脸,训练时不容易收敛,如图:
mask

(5) 利用data/FACE文件中的脚本,把VOC格式转换成LMDB格式,如图:
lmdb

在caffe/data目录下创建faces_database文件夹,拷贝wider_face文件夹(前面生成的VOC格式数据),layout如图:
database

cd caffe
\# Create the trainval.txt, test.txt, and test_name_size.txt in data/FACE/                
./data/FACE/create_list.sh                                                                
\# You can modify the parameters in create_data.sh if needed.                             
\# It will create lmdb files for trainval and test with encoded original image:           
\#   data/faces_database/FACE/lmdb/FACE_trainval_lmdb                                     
\#   data/faces_database/FACE/lmdb/FACE_test_lmdb                                         
\# and make soft links at examples/FACE/                                                  
./data/FACE/create_data.sh                                                                

3. 训练:


训练需要的参数文件和网络文件位置如图: train

运行以下命令开始训练:
./build/tools/caffe train --solver examples/faceboxes/solver.prototxt

4. 测评:


测评脚本以及模型文件位置如图: test

FDDB上的测评结果(discontinuous)如图:
result

论文中的结果:
origin

效果图:
demo

5. 参考:


参考的仓库:https://github.com/lsy17096535/faceboxes

6. 优化:


train.prototxt支持Anchor densification strategy

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].