几个不错深度学习github源码（人体姿态、物体姿态、人脸相关、行人重识别、行人属性、风格迁移、目标检测、OCR）

搜索了下github，把觉得几个不错的深度学习代码整理下列出来，未来可能会去涉及这些。包括人体姿态、物体姿态、人脸相关、行人重识别、行人属性、风格迁移、目标检测。主要以pytorch或tensorflow为主。

 

目录

人体姿态估计

物体姿态估计

人脸、其他

行人重识别

行人属性

风格迁移

目标检测

OCR

---

人体姿态估计

• Deep Pose Estimation implemented using Tensorflow with Custom Architectures for fast inference.
  • code：https://github.com/ildoonet/tf-pose-estimation.git
• pytorch_Realtime_Multi-Person_Pose_Estimation
  • code：https://github.com/tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation.git
• Code repo for realtime multi-person pose estimation in CVPR'17 (Oral)
  • code：https://github.com/ZheC/Realtime_Multi-Person_Pose_Estimation.git
• Light-weight Single Person Pose Estimator
  • code: https://github.com/Smallflyfly/MobilePose-pytorch.git

物体姿态估计

• This research project implements a real-time object detection and pose estimation method as described in the paper, Tekin et al. "Real-Time Seamless Single Shot 6D Object Pose Prediction", CVPR 2018.
  • code: https://github.com/microsoft/singleshotpose.git

人脸、其他

• 一款入门级的人脸、视频、文字检测以及识别的项目.
  • code: https://github.com/vipstone/faceai.git
• papers about Face Detection; Face Alignment; Face Recognition && Face Identification && Face Verification && Face Representation; Face Reconstruction; Face Tracking; Face Super-Resolution && Face Deblurring; Face Generation && Face Synthesis; Face Transfer; Face Anti-Spoofing; Face Retrieval
  • code: https://github.com/ChanChiChoi/awesome-Face_Recognition.git
• A MNIST-like fashion product database. Benchmark
  • code：https://github.com/zalandoresearch/fashion-mnist.git
• The world's simplest facial recognition api for Python and the command line
  • code: https://github.com/ageitgey/face_recognition.git
• Face recognition using Tensorflow
  • code: https://github.com/davidsandberg/facenet.git
• The world's simplest facial recognition api for Python and the command line
  • code: https://github.com/ageitgey/face_recognition.git

行人重识别

• Torchreid: Deep learning person re-identification in PyTorch.
  • code: https://github.com/KaiyangZhou/deep-person-reid.git
• A tiny, friendly, strong pytorch implement of person re-identification baseline.
  • code: https://github.com/layumi/Person_reID_baseline_pytorch.git
• Bag of Tricks and A Strong Baseline for Deep Person Re-identification
  • code: https://github.com/michuanhaohao/reid-strong-baseline.git
• MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
  • code: https://github.com/Xiaoccer/MobileFaceNet_Pytorch.git

行人属性

• A simple baseline implemented in PyTorch for pedestrian attribute recognition task, evaluating on Market-1501 and DukeMTMC-reID dataset.
  • code: https://github.com/hyk1996/Person-Attribute-Recognition-MarketDuke.git
• The paper list of person attribute recognition
  • code: https://github.com/wangxiao5791509/Pedestrian-Attribute-Recognition-Paper-List.git

风格迁移

• Neural style in TensorFlow! 
  • code: https://github.com/Smallflyfly/neural-style.git
• Neural Style and MSG-Net
  • code: https://github.com/Smallflyfly/PyTorch-Multi-Style-Transfer.git

目标检测

• Faster R-CNN
  • code: https://github.com/jwyang/faster-rcnn.pytorch.git
• YoloV3 YOLOv3 in PyTorch > ONNX > CoreML > iOS
  • code: https://github.com/ultralytics/yolov3.git
• YOLOv3
  • code: https://github.com/eriklindernoren/PyTorch-YOLOv3.git
• SSD
  • code: https://github.com/amdegroot/ssd.pytorch.git

OCR

• yolo3+ocr

  • code: https://github.com/chineseocr/chineseocr.git

• 运用tf实现自然场景文字检测,keras/pytorch实现ctpn+crnn+ctc实现不定长场景文字OCR识别
  • code: https://github.com/xiaofengShi/CHINESE-OCR.git

• Official implementation of Character Region Awareness for Text Detection (CRAFT)

  • code: https://github.com/clovaai/CRAFT-pytorch.git

• DataFountain HUAWEI OCR 2019 rank Top1% ,Text Localization using pixellink,Word Recognition using densenet
  • code: https://github.com/HLIG/HUAWEIOCR-2019.git