基于opencv的人体姿态估计,实现视频的4分类

上传者: 45107081 | 上传时间: 2023-03-14 10:33:01 | 文件大小: 794.28MB | 文件类型: RAR
本设计将基于OpenCV,采用“关键点提取并归一化”与“分类器”相结合的方式,实现多人正常和异常姿态识别的设计。关键词 OpenCV 人体姿态估计 多分类 行为识别;主要功能是通过MoveNet对前期用于训练的视频内容进行人体骨骼关键点信息的提取,MoveNet将在每帧上将人体骨骼关键点的x和y坐标提取出来,通过一定的算法进行归一化,并保存数据。首先,通过OpenCV将视频读取,通过OpenCV进行简单的视频预处理,进行BGR转RGB的操作,然后加载MoveNet的关键点模型将人体骨骼关键点信息提取出来,对每帧的x和y的坐标进行归一化,将不同大小的骨骼标准化,按帧存入数据库中,这个过程将按WALK、STAND、FALL、FIGHT这四类动作分别进行提取与处理。主要功能对前期数据库内容进行数据分割,生成4个LSTM模型,对分割好的数据进行导入,并和导入对应标签进行迭代训练,最后生成Loss值最低的模型。主要功能对前期数据库内容进行数据分割,生成4个LSTM模型,对分割好的数据进行导入,并和导入对应标签进行迭代训练,最后生成Loss值最低的模型。

文件下载

资源详情

[{"title":"( 43 个子文件 794.28MB ) 基于opencv的人体姿态估计,实现视频的4分类","children":[{"title":"poose","children":[{"title":".vs","children":[{"title":"ProjectSettings.json <span style='color:#111;'> 37B </span>","children":null,"spread":false},{"title":"slnx.sqlite <span style='color:#111;'> 76.00KB </span>","children":null,"spread":false},{"title":"VSWorkspaceState.json <span style='color:#111;'> 209B </span>","children":null,"spread":false},{"title":"poose","children":[{"title":"v15","children":[{"title":".suo <span style='color:#111;'> 20.50KB </span>","children":null,"spread":false}],"spread":true},{"title":"v16","children":[{"title":".suo <span style='color:#111;'> 51.50KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}],"spread":true},{"title":"poose","children":[{"title":"Tracker.py <span style='color:#111;'> 9.36KB </span>","children":null,"spread":false},{"title":"fall_fight_waalk.mp4 <span style='color:#111;'> 31.86MB </span>","children":null,"spread":false},{"title":"trainy.txt <span style='color:#111;'> 1.59KB </span>","children":null,"spread":false},{"title":"trainx-Copy1.txt <span style='color:#111;'> 3.93MB </span>","children":null,"spread":false},{"title":"photo.jpg <span style='color:#111;'> 156.14KB </span>","children":null,"spread":false},{"title":"dance.mp4 <span style='color:#111;'> 4.41MB </span>","children":null,"spread":false},{"title":"fight1.mp4 <span style='color:#111;'> 135.36MB </span>","children":null,"spread":false},{"title":"fall1.mp4 <span style='color:#111;'> 76.53MB </span>","children":null,"spread":false},{"title":"my_best_model.epoch{epoch_02d}-loss{val_loss_.2f}.hdf5 <span style='color:#111;'> 981.04KB </span>","children":null,"spread":false},{"title":"dance1.mp4 <span style='color:#111;'> 208.53MB </span>","children":null,"spread":false},{"title":"trainy-Copy1.txt <span style='color:#111;'> 1.59KB </span>","children":null,"spread":false},{"title":"stand1.mp4 <span style='color:#111;'> 93.13MB </span>","children":null,"spread":false},{"title":"хоЙшгЕцХЩчиЛ.docx <span style='color:#111;'> 663.17KB </span>","children":null,"spread":false},{"title":"FPS.py <span style='color:#111;'> 1.21KB </span>","children":null,"spread":false},{"title":"poose.ipynb <span style='color:#111;'> 58.31KB </span>","children":null,"spread":false},{"title":"models","children":[{"title":"movenet_multipose_lightning_192x256_FP32.bin <span style='color:#111;'> 18.04MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_320x320_FP32.bin <span style='color:#111;'> 18.07MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_256x256_FP32.xml <span style='color:#111;'> 515.12KB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_480x640_FP32.xml <span style='color:#111;'> 515.78KB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_192x192_FP32.bin <span style='color:#111;'> 18.04MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_256x320_FP32.bin <span style='color:#111;'> 18.06MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_256x256_FP32.bin <span style='color:#111;'> 18.05MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_320x320_FP32.xml <span style='color:#111;'> 515.28KB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_480x640_FP32.bin <span style='color:#111;'> 18.17MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_192x256_FP32.xml <span style='color:#111;'> 515.12KB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_736x1280_FP32.bin <span style='color:#111;'> 18.47MB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_256x320_FP32.xml <span style='color:#111;'> 515.19KB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_192x192_FP32.xml <span style='color:#111;'> 515.09KB </span>","children":null,"spread":false},{"title":"movenet_multipose_lightning_736x1280_FP32.xml <span style='color:#111;'> 515.86KB </span>","children":null,"spread":false}],"spread":false},{"title":"trainx.txt <span style='color:#111;'> 3.93MB </span>","children":null,"spread":false},{"title":"my_best_model-Copy1.epoch{epoch_02d}-loss{val_loss_.2f}.hdf5 <span style='color:#111;'> 981.04KB </span>","children":null,"spread":false},{"title":"__pycache__","children":[{"title":"FPS.cpython-38.pyc <span style='color:#111;'> 1.72KB </span>","children":null,"spread":false},{"title":"Tracker.cpython-38.pyc <span style='color:#111;'> 8.17KB </span>","children":null,"spread":false},{"title":"FPS.cpython-310.pyc <span style='color:#111;'> 1.74KB </span>","children":null,"spread":false}],"spread":false},{"title":"MovenetOpenvino.py <span style='color:#111;'> 15.66KB </span>","children":null,"spread":false},{"title":".ipynb_checkpoints","children":[{"title":"poose-checkpoint.ipynb <span style='color:#111;'> 1.66MB </span>","children":null,"spread":false}],"spread":false},{"title":"output.mp4 <span style='color:#111;'> 64.38MB </span>","children":null,"spread":false},{"title":"walk1.mp4 <span style='color:#111;'> 63.65MB </span>","children":null,"spread":false}],"spread":false}],"spread":true}],"spread":true}]

评论信息

免责申明

【只为小站】的资源来自网友分享,仅供学习研究,请务必在下载后24小时内给予删除,不得用于其他任何用途,否则后果自负。基于互联网的特殊性,【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查;无论 【只为小站】 经营者是否已进行审查,用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场,基于网友分享,根据中国法律《信息网络传播权保护条例》第二十二条之规定,若资源存在侵权或相关问题请联系本站客服人员,zhiweidada#qq.com,请把#换成@,本站将给予最大的支持与配合,做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明