搜索资源列表
Intelligent_Screen_Saver
- It is a utility to control screen saver using human face detection. Human face detection is performed using OpenCV Haar-Cascade method. The software is primarily a daemon that resides in your system tray and keeps observing the input from a webcam to
poppe_humaneva_code.zip
- 人体检测matlab程序。主要用来处理humanEva数据集。,The code is mainly to evaluate the Example-based Pose Estimation on the HumanEva sets. This work is published in CVPR – EHuM workshop, June 2007.
Paper]
- 最新顶级会议CVPR2011关于人体姿态识别文章-Paper] Real-time Human Pose Recognition in Parts from Single Depth Images 2011CVPRBodyPartRecognition
ramanan-cvpr05_1
- 斗殴行为识别 Strike a Pose: Tracking People by Finding Stylized Poses-Fight acts of identification Strike a Pose: Tracking People by Finding Stylized Poses
FaceDetection_Based_on_a_New_Nonlinear_Color_Space
- 提出一种新的非线性变换的彩色空间 ″″, 利用次高斯概率分布函数拟合皮肤色度信息, 得到候选区 YC C r b 域。为了排除候选区域中的非人脸, 首先根据均值和方差信息分割出候选区域中的纹理特征信息, 再通过多尺度 ) ( 信息定位眼睛, 然后根据人脸特征的几 形态边缘检测算子检测候选区域的边缘, 利用 边缘方向 PCA PCAED ( ) 何形状信息定位其他特征 鼻、嘴 , 通过这些几何特征信息对肤色分割得到的候选区域进行验证, 最终得到正确 的人
llr
- 基于LLR算法的多姿态人脸识别，是对以往的改进版本-LLR-based Multi-pose face recognition algorithm is an improved version of previous
Adaboost
- 基于Adaboost算法的多姿态人脸实时视频检测江苏大学-Adaboost algorithm based on multi-pose face detection real-time video
1
- 一种处理多姿态人脸识别的多候选类加权识别方法-Weighted identification method for a treatment of multi-candidate class of multi-pose face recognition
IMproved-LBP
- 基于大尺度纹理特征描述算子，对图像纹理特征描述性更好，更好的把握图像的粗粒特征，对光照姿态表情变化的鲁棒性更好-Based on large-scale texture descr iption operator, descr iption of the image texture features better and better grasp the coarse image characteristic changes in the expression for illumination
Using-Stereo-Matching-for-2-D-Face-Recognition-Ac
- Using Stereo Matching for 2-D Face Recognition Across Pose
Estimation
- Program for estimation of head pose
mfc_pictureShow
- 使用MFC实现的双目视觉里程计离线编程项目，能计算出机器人行走位姿-Use binocular vision odometer offline programming project MFC implementation, and can calculate the robot walking pose
accv09_pose
- 从二维图像恢复三维的人体姿势，即便发生部分或者严重的人体遮挡。-We address the problem of recovering 3D human pose single 2D images when humans are partially or heavily occluded in the scenes
tf-pose-estimation-master
- OpenPose人体姿态识别项目是美国卡耐基梅隆大学（CMU）基于卷积神经网络和监督学习并以caffe为框架开发的开源库。可以实现人体动作、面部表情、手指运动等姿态估计。适用于单人和多人，具有极好的鲁棒性。是世界上首个基于深度学习的实时多人二维姿态估计应用，基于它的实例如雨后春笋般涌现。人体姿态估计技术在体育健身、动作采集、3D试衣、舆情监测等领域具有广阔的应用前景，人们更加熟悉的应用就是抖音尬舞机(OpenPost Human Attitude Recognition Project is a
mediapipe
- 手势识别,基于mediapipe,可支持pose、hand、face(Gesture recognition, based on mediapipe, can support pose, hand, face)