專案地址：GitHub - stereolabs/zed-examples: ZED SDK Example projects
官方檔案：Stereolabs Docs: API Reference, Tutorials, and Integration
How to Use PyTorch with ZED | Stereolabs
GitHub - stereolabs/zed-pytorch: 3D Object detection using the ZED and Pytorch
https://github.com/stereolabs/zed-pytorch
3D Mask R-CNN using the ZED and Pytorch
The ZED SDK can be interfaced with Pytorch for adding 3D localization of custom objects detected with Mask-RCNN. In this Python 3 sample, we will show you how to detect, segmente, classify and locate objects in 3D space using the ZED stereo camera and Pytorch.
1. 環境配置
雙目立體視覺（9）PyTorch & ZED 3D人體識別與追蹤 (上)_Techblog of HaoWANG-CSDN博客專案地址：GitHub - stereolabs/zed-examples: ZED SDK Example projects官方檔案：Stereolabs Docs: API Reference, Tutorials, and IntegrationHow to Use PyTorch with ZED | Stereolabshttps://www.stereolabs.com/docs/pytorch/github專案地址：3D Mask R-CNN using the ZED and ...https://haowang.blog.csdn.net/article/details/121227055 在安裝torch和編譯maskrcnn時，不同的環境配置會有例外或者報錯，參考上一篇博客解決，
2. 測驗驗證
從這個存盤庫下載示例代碼，我們提供了一個簡單的網路攝像機演示，演示了如何使用 maskrcnn 進行推理:
激活conda環境：
conda activate pytorch_env下載預訓練模型，測驗運行：
python zed_object_detection.py --config-file configs/caffe2/e2e_mask_rcnn_R_50_C4_1x_caffe2.yaml --min-image-size 256
注意：顯卡的顯存不夠時會報錯
也開始使用其他版本：
# A SVO file can be loaded
python zed_object_detection.py --svo-filename path/to/svo_file.svo
# for best results, use min-image-size 800
python zed_object_detection.py --min-image-size 800
# or change the model that you want to use
python zed_object_detection.py --config-file configs/caffe2/e2e_mask_rcnn_R_101_FPN_1x_caffe2.yaml --min-image-size 300
# in order to see the probability heatmaps, pass --show-mask-heatmaps
python zed_object_detection.py --min-image-size 300 --show-mask-heatmaps
# for the keypoint demo
python zed_object_detection.py --config-file configs/caffe2/e2e_keypoint_rcnn_R_50_FPN_1x_caffe2.yaml --min-image-size 300
# can also run it on the CPU
python zed_object_detection.py --min-image-size 300 MODEL.DEVICE cpu
模型會被自動下載，如果等待時間過長，請直接在命令列內右單擊下載網址，手動下載并重命名與模型一致，且置于指定檔案夾中，重新運行即可跳過自動下載步驟，
使用CPU版本效果展示：
轉載請註明出處，本文鏈接：https://www.uj5u.com/qita/356711.html
標籤：AI