nianticlabs/manydepth issues and pull requests

#77 - freeze_teacher

Issue - State: open - Opened by 757787182 7 months ago

#76 - test-time refinement code

Issue - State: open - Opened by Zhangwenyao1 12 months ago

#75 - custom dataset prepare

Issue - State: open - Opened by 1991yuyang about 1 year ago

#74 - On the pre training model of cityscapes

Issue - State: open - Opened by WangXuCh over 1 year ago - 3 comments

#73 - How to conduct multi GPU training

Issue - State: open - Opened by WangXuCh over 1 year ago

#72 - what should the

Issue - State: closed - Opened by Gxhappiness over 1 year ago

#71 - The dataset GT-Calculation

Issue - State: closed - Opened by wangjiyuan9 almost 2 years ago - 1 comment

#70 - About qualitative results presented in Figure 4

Issue - State: open - Opened by YagumaLi almost 2 years ago - 1 comment

#69 - Calculate the gt depth for Cityscape images

Issue - State: closed - Opened by wangjiyuan9 almost 2 years ago - 3 comments

#68 - Get gt depth for other Cityscape images

Issue - State: closed - Opened by wangjiyuan9 almost 2 years ago - 2 comments

#67 - About MAX gt_depth

Issue - State: closed - Opened by wangjiyuan9 almost 2 years ago - 2 comments

#66 - when i test the image,the result of the depth is black

Issue - State: open - Opened by 9796l about 2 years ago - 1 comment

#65 - Why the training time of manydepth is much shorter than monodepth2?

Issue - State: closed - Opened by myalos about 2 years ago - 3 comments

#64 - About scale problem in monocular setting

Issue - State: open - Opened by myalos about 2 years ago - 2 comments

#63 - Help training out bright reflection...

Issue - State: open - Opened by pgaston about 2 years ago - 1 comment

#62 - how to test many frames at the same time?

Issue - State: open - Opened by 9796l about 2 years ago

#61 - Question of relative pose in matching augmentation.

Issue - State: open - Opened by JarvisLee0423 over 2 years ago - 1 comment

#60 - Depth Estimation Results on Single Frames

Issue - State: open - Opened by ldkong1205 over 2 years ago

#59 - Can the scale ambiguity be resolved with multi-frame approach?

Issue - State: open - Opened by JinraeKim over 2 years ago - 1 comment

#58 - Add test_simple_multiple.py

Pull Request - State: closed - Opened by JinraeKim over 2 years ago - 1 comment

#57 - Update evaluate_depth.py

Pull Request - State: open - Opened by aartykov over 2 years ago

#56 - Change from Image.ANTIALIAS to transforms.InterpolationMode.LANCZOS

Pull Request - State: open - Opened by aartykov over 2 years ago

#55 - About cityscapes disparity to depth_gt

Issue - State: open - Opened by willemeng over 2 years ago - 2 comments

#54 - Stereo + Tempora (training with stereo video like monodepth2)

Issue - State: open - Opened by Ecalpal over 2 years ago

#53 - Training on NYU-V2 Dataset

Issue - State: open - Opened by mshaheryarmalik over 2 years ago

#52 - The results of cityscapes

Issue - State: closed - Opened by Ecalpal over 2 years ago - 3 comments

#51 - Depth evaluation for single frame mode?

Issue - State: closed - Opened by LeeBY68 almost 3 years ago - 2 comments

#50 - Depth estimation from underwater monocular video sequences

Issue - State: closed - Opened by buster-zbb almost 3 years ago - 4 comments

#49 - Bug in intrinsics re-scaling?

Issue - State: closed - Opened by aj96 almost 3 years ago - 2 comments

#48 - inconsistent and incorrect scale in predicted depth

Issue - State: open - Opened by ChauChorHim almost 3 years ago - 5 comments

#47 - parameters of the model

Issue - State: closed - Opened by xiaonan320 almost 3 years ago

#46 - A tiny bug during training

Issue - State: open - Opened by ZhanyuGuo almost 3 years ago

#45 - how to get the RGBD map like the video shows

Issue - State: open - Opened by heroacool almost 3 years ago - 2 comments

#44 - Can't get custom dataset to perform as well as monodepth2

Issue - State: open - Opened by d4l3k about 3 years ago - 2 comments

#43 - how to choose the best freeze_teacher_epoch parameter?

Issue - State: closed - Opened by dkerjw1 over 3 years ago

#42 - Question About Input Image Size

Issue - State: closed - Opened by benhgm over 3 years ago - 1 comment

#41 - Input channels for input_features for PoseDecoder

Issue - State: open - Opened by benhgm over 3 years ago

#40 - Indoor environment

Issue - State: open - Opened by suyuzhang over 3 years ago - 9 comments

#39 - About monodepth teacher

Issue - State: closed - Opened by biggiantpigeon over 3 years ago

#38 - How to get the error map between predicted depth and the GT depth map

Issue - State: open - Opened by ruili3 over 3 years ago - 1 comment

#37 - Pip module

Pull Request - State: open - Opened by AdityaNG over 3 years ago - 1 comment

#36 - depth ground truth error

Issue - State: open - Opened by seb-le over 3 years ago

#35 - Can't reproduce the results on cityscapes

Issue - State: open - Opened by agenthong over 3 years ago - 28 comments

#34 - Moving object from left to right

Issue - State: closed - Opened by biggiantpigeon over 3 years ago - 1 comment

#33 - about confidence_mask

Issue - State: open - Opened by biggiantpigeon over 3 years ago - 2 comments

#32 - Hi, I would like to make two small suggestions about the evaluate code

Issue - State: open - Opened by Ecalpal over 3 years ago - 2 comments

#31 - What should cityscapes looks like?

Issue - State: closed - Opened by agenthong over 3 years ago

#30 - About update_adaptive_depth_bins in trainer.py

Issue - State: open - Opened by ccj5351 over 3 years ago - 2 comments

#29 - Custom data sets (high speed scenes) do not work well

Issue - State: open - Opened by Klutch0615 over 3 years ago - 1 comment

#28 - If files are missing in CS eval, no more silent failures

Pull Request - State: closed - Opened by mdfirman over 3 years ago

#27 - About test-time-refinement (TTR)

Issue - State: closed - Opened by heiwang1997 over 3 years ago - 5 comments

#26 - pose_enc.load_state_dict error

Issue - State: open - Opened by boyagesmile over 3 years ago

#25 - Freeze Teacher network from beginning

Issue - State: closed - Opened by pjckoch over 3 years ago - 2 comments

#24 - How to debug when training with "python -m manydepth.train"

Issue - State: closed - Opened by ruili3 over 3 years ago - 1 comment

#23 - How can I save predicted depth map?

Issue - State: closed - Opened by agenthong over 3 years ago - 3 comments

#22 - Wrong depth scale when using ground-truth camera poses.

Issue - State: open - Opened by ootts over 3 years ago - 3 comments

#21 - how to do the ablation study without teacher

Issue - State: closed - Opened by sunnyHelen over 3 years ago - 1 comment

#20 - AssertionError: size of input tensor and input format are different.

Issue - State: closed - Opened by agenthong over 3 years ago - 2 comments

#19 - bug fix

Pull Request - State: closed - Opened by frkake over 3 years ago - 3 comments

#18 - batch size 8 got abs rel 0.130

Issue - State: closed - Opened by sunnyHelen over 3 years ago - 5 comments

#17 - Why disable gradients of on lookup images?

Issue - State: closed - Opened by WBS-123 over 3 years ago - 7 comments

#16 - How much memory does this cost?

Issue - State: closed - Opened by agenthong over 3 years ago - 4 comments

#15 - Pre-trained models for monocular depth networks

Issue - State: closed - Opened by VitorGuizilini-TRI over 3 years ago - 5 comments

#14 - Can you provide more ablations about mask?

Issue - State: closed - Opened by whyygug over 3 years ago - 2 comments

#13 - Training with Lyft

Issue - State: closed - Opened by didriksg over 3 years ago - 8 comments

#12 - Stereo + Temporal

Issue - State: closed - Opened by fengziyue over 3 years ago - 2 comments

#11 - Depth map scale for KITTI data

Issue - State: closed - Opened by Benjabby almost 4 years ago - 5 comments

#10 - Feature request: Tensorflow lite version

Issue - State: closed - Opened by rodrigoGA almost 4 years ago - 1 comment

#9 - get_depth never enabled

Issue - State: open - Opened by didriksg almost 4 years ago - 3 comments

#8 - Train on own dataset with not good result

Issue - State: closed - Opened by xuzhiyuan29 almost 4 years ago - 5 comments

#7 - Detach the reprojection loss mask needed?

Issue - State: closed - Opened by DaoyiG almost 4 years ago - 1 comment

#6 - About depth bins

Issue - State: closed - Opened by hyBlue almost 4 years ago - 2 comments

#5 - Contact person / email

Issue - State: closed - Opened by Yevkuzn almost 4 years ago - 2 comments

#4 - "Normal" Training Loss and Strange Test Result

Issue - State: closed - Opened by Owen-Liuyuxuan almost 4 years ago - 4 comments

#3 - Strange training loss and test result

Issue - State: closed - Opened by Owen-Liuyuxuan almost 4 years ago - 3 comments

#2 - Multi-GPU

Issue - State: closed - Opened by fengziyue almost 4 years ago - 2 comments

#1 - Supplementary materials

Issue - State: closed - Opened by fengziyue almost 4 years ago - 2 comments

GitHub / nianticlabs/manydepth issues and pull requests