iResNet icon indicating copy to clipboard operation
iResNet copied to clipboard

Results 16 iResNet issues
Sort by recently updated
recently updated
newest added

Hi,I have not found the prototxt about training the ROB SceneFlow,Can you provide the corresponding prototxt? In this project,4 dataset was used in the train_ROB_stage_one.prototxt,and 3 dataset was used in...

Hi, While training the network on my own dataset, the losses ires_disp_loss0, ires_disp_loss1 and ires_disp_loss2 looks same. Is this seem reasonable ? ![image](https://user-images.githubusercontent.com/40571095/56959414-38b7dd00-6b45-11e9-95f6-e59b1b780c96.png)

我使用预训练模型iResNet_ROB.caffemodel作为程序的输入参数,我想直接测试kitti中的数据,所以我的主函数为: if __name__ == '__main__': caffemodel = sys.argv[1] training_dataset_dir = os.path.join(dataset_dir,'training') test_dataset_dir = os.path.join(dataset_dir,'test') # test kitti dataset_name = 'Kitti2015' # test set RunMethod(caffemodel, submit_dir, test_dataset_dir, dataset_name, False) 当我执行到im0_size =...

hello, thank you for the helpful installation instructions, but i'm stuck at the data preparation phase, because the following link seems to be broken for me: http://www.cvlibs.net:3000/ageiger/rob_devkit how should i...

python ../train_rob.py 2>&1 | tee rob.log F0610 14:24:18.777849 8200 io.cpp:36] Check failed: fd != -1 (-1 vs. -1) File not found: ../ROB_training/solver_rob_stage_one.prototxt *** Check failure stack trace: *** @ 0x7fc5b12285cd...

python ../train_rob.py 2>&1 | tee rob.log ![2](https://user-images.githubusercontent.com/20469179/48194484-22944580-e388-11e8-9c1b-096c789c9b02.png) Hi, when I run "python ../train_rob.py 2>&1 | tee rob.log",I ran into the problem above , can you help me?Thank you very much!

I want to test on CVPR version (SceneFlow + KITTI). I tried to perform test_rob.py using the model(deploy_iresnet.tpl.prototxt) and the pretrained Model(CVPR2018 KITTI 2015), but result didn't look correct. Please...

First of all. Thank you for sharing the code. I have some questions for training scheme which is confused with training prototxt. In the [https://github.com/leonzfa/iResNet/blob/master/models/ROB_training/train_rob_stage_one.prototxt](url), iResNet is trained with four...

I'm running your code on the KITTI dataset, but the runtime for processing a single image pair is around 0.35s instead of 0.12s as stated in the paper. In order...