UniMVSNet
UniMVSNet copied to clipboard
prstrive
[CVPR 2022] Rethinking Depth Estimation for Multi-View Stereo: A Unified Representation
Hi, you present the point clouds based on gipuma fusion, but I got some problems when I tried to install fusible. Could please me the environment you applied, including the...
I tried to infer and fuse on DTU dataset, but found the predicted results are different from your released ply models. For example: Your released point cloud for scan1 (mvsnet001_l3.ply):...
谢谢大佬分享那么好的成果，想再请教一个问题，我看代码里设置的fusion参数 '--prob_threshold', '0.3', \ '--disp_threshold', '0.25', \ '--num_consistent', '3'] 然后我看CVP-MVSNet里设置的如下，这个一般是有什么依据的？ parser.add_argument('--prob_threshold', type=float, default = '0.8') parser.add_argument('--disp_threshold', type=float, default = '0.13') parser.add_argument('--num_consistent', type=float, default = '3')
关于小数据集训练
感谢大佬分享这个很棒的工作，我用自己采集的数据，大约400张，用lidar做真值的，用blendedmvs_finetune.sh这个脚本去从头开始训练，大约30个epoch，可以收敛，但是预测不出来结果或者只有很少的点。 然后从BlendedMVS数据集中抽了其中一个场景来做同样的实验，也是预测不出来结果。 如果用整个BlendedMVS数据集来从头开始训练，不加载预训练模型，预测结果是正常的。 初始学习率设的是0.001。 DTU数据集部分的代码我也看了，就是同一个视角拍了七张不同打光的照片，其他基本上跟BlendedMVS数据集是一样的。 百思不得其解，想请教下大佬为什么小数据集训练会有问题？
I have a problem on the tanks dataset: for all 8 scenes, the private testing result is 79.61 65.66 54.16 59.51 62.24 60.90 57.61 54.59 i follow the author's resconstruction...
Thanks for your generous open source. I just wanna know if the weights you offered is only for the "fpn". When I turn the "fea_mode" to "unet", there are errors...
Hello. Could you please tell me how you confirm the 3-step conf values of different classifications for the 'Tanks and Temples' dataset? Do I need use another neural network to...
使用fusibile进行点云的融合时，报以下的错误：Found 0.00 million points，是什么原因呢 Fusing points Processing camera 0 Found 0.00 million points Processing camera 1 Found 0.00 million points Processing camera 2 Found 0.00 million points Processing camera 3...
Hello, thank you very much for your open source. I used the mobile phone to collect data and use colmap sparse reconstruction to get the camera parameters, and then use...
