定制網(wǎng)站需要多少錢,wordpress添加百度地圖,鄭州網(wǎng)站制作公司哪家好,視頻推廣方案目錄 清華GUAVA#xff1a;一張照片就能搞定3D人#xff0c;速度1分鐘左右#xff0c;流程比較長 EHM-Tracker 模型#xff1a;stylematte_synth.pt 清華GUAVA#xff1a;一張照片就能搞定3D人#xff0c;速度1分鐘左右#xff0c;流程比較長 論文地址: https://arxi…目錄清華GUAVA一張照片就能搞定3D人速度1分鐘左右流程比較長EHM-Tracker模型stylematte_synth.pt清華GUAVA一張照片就能搞定3D人速度1分鐘左右流程比較長論文地址: https://arxiv.org/pdf/2505.03351項目主頁https://eastbeanzhang.github.io/GUAVA/開源代碼https://github.com/Pixel-Talk/GUAVA視頻Demo: https://www.bilibili.com/video/BV1k45AzaEpW/依賴項pytorch3dhttps://blog.csdn.net/jacke121/article/details/143951830下載username$(urle $username) password$(urle $password) wget --post-data username$usernamepassword$password https://download.is.tue.mpg.de/download.php?domainsmplxsfileSMPLX_NEUTRAL_2020.npzresume1 -O ./data/HPS/pixie_data/SMPLX_NEUTRAL_2020.npz --no-check-certificate --continue # PIXIE pretrained model and utilities wget --post-data username$usernamepassword$password https://download.is.tue.mpg.de/download.php?domainpixiesfilepixie_model.tarresume1 -O ./data/HPS/pixie_data/pixie_model.tar --no-check-certificate --continue wget --post-data username$usernamepassword$password https://download.is.tue.mpg.de/download.php?domainpixiesfileutilities.zipresume1 -O ./data/HPS/pixie_data/utilities.zip --no-check-certificate --continue cd data/HPS/pixie_dataEHM-Tracker算法功能半身動捕python tracking_video.py -i /tmp/gradio/99d38740e32b7465bd5274ccf3b054650b546d9061e5b9913a1387cd13dcd0ff/b002.mp4 -o /data/lbg/project/GUAVA-main/outputs/app/tracked_driven_video --check_hand_score 0.0 -p 0,1 -n 1 -v 0Working directory: /data/lbg/project/GUAVA-main/EHM-Tracker模型stylematte_synth.pt這個算法提供了模型包好多模型都有dwpose fan hamer lmk106 lmk70 matting mediapipe pixie teaser vposertarget: src.modules.human_matting.matting.StyleMatteModel params: ckpt_fp: pretrained/matting/stylematte_synth.pt device: cuda:0下載wget https://github.com/chroneus/stylematte/releases/download/weights/stylematte_synth.pthcd /data/lbg/project/GUAVA-main/EHM-Tracker python tracking_video.py -i /data/lbg/project/GUAVA-main/data/nan_5s.mp4 -o /data/lbg/project/GUAVA-main/outputs/app/tracked_driven_video --check_hand_score 0.0 -p 0,1 -n 1 -v 0 ffmpeg -i nan1.mp4 -t 8 -c:v libx264 -c:a aac nan_5s.mp4