WangYueFt/dgcnn issues and pull requests

#88 - About spatial transform module

Issue - State: open - Opened by Howtyee 9 days ago

#87 - CUDA error

Issue - State: open - Opened by xiaohuihuifirst 5 months ago

#86 - 数据可视化严重问题

Issue - State: open - Opened by zhangzhenyu-zzy 10 months ago

#85 - The network architecture in code is different from the paper?

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

#84 - What is the KNN distance formula implemented by pytorch

Issue - State: open - Opened by zyt1024 over 1 year ago - 1 comment

#83 - ValueError: need at least one array to concatenate

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

#82 - Visualization

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

#81 - Why is the accuracy rate so low?

Issue - State: open - Opened by sunhufei almost 3 years ago - 1 comment

#80 - Is the input size of the first layer 6 because the input parameters are the point itself and its normals??

Issue - State: closed - Opened by FandosA almost 3 years ago - 1 comment

#79 - Aborted (core dumped) if I process to many points at once

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

#78 - How to add more DGCNN layers in your implementation?

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

#77 - Why does the pytorch version not have transform？

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

#76 - LiDAR Point Cloud Classification results not good with real data

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

#75 - 请问你的分类模型到底是哪样的，有的有连接，有的没有！！！

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

#74 - Understanding equation #8 in research paper (Section 3.1)

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

#73 - Open3d 090

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

#72 - Compatibility with tensorflow 2, python3.8

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

#71 - getting 'Indexerror' in train_multi_gpu.py

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

#70 - Visualization issues

Issue - State: open - Opened by guocx1g over 3 years ago

#69 - Can we run this code in windows?

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

#68 - pytorch part segmentation

Issue - State: open - Opened by sank96 over 3 years ago - 1 comment

#67 - Can I get the final paper?

Issue - State: open - Opened by AlbertoWang almost 4 years ago

#66 - There was no model.2048.t7 file in pretrained folder.

Issue - State: closed - Opened by byliut almost 4 years ago

#65 - How to implement 92.9% OA?

Issue - State: open - Opened by 357998846 over 4 years ago - 4 comments

#64 - how to set the value of K with different points for other datasets

Issue - State: open - Opened by longmalongma over 4 years ago

#63 - Questions about python and tensorFlow versions

Issue - State: open - Opened by Noahjianlong over 4 years ago

#62 - tensorflow/provider.py

Issue - State: closed - Opened by NayeeC over 4 years ago - 1 comment

#61 - Semantic segmentation: Result of 4D input is worse than 3D input

Issue - State: open - Opened by MCLYang over 4 years ago - 1 comment

#60 - What effect did you expect by considering 'categorical vector'?

Issue - State: open - Opened by hjsg1010 over 4 years ago

#59 - How did you calculate forward time for several models?

Issue - State: closed - Opened by hjsg1010 over 4 years ago - 3 comments

#58 - How do you visualize your segmentation outputs?

Issue - State: closed - Opened by hjsg1010 over 4 years ago - 3 comments

#57 - python collect_indoor3d_data.py ERROR

Issue - State: closed - Opened by hjsg1010 over 4 years ago - 1 comment

#56 - Potential discrepancy between training and testing for part segmentation

Issue - State: closed - Opened by imankgoyal over 4 years ago - 5 comments

#55 - Call contiguous() after permute() in the pytorch model

Pull Request - State: closed - Opened by hqucms over 4 years ago

#54 - Can you provide a pre-trained model for semantic segmentation?

Issue - State: closed - Opened by longmalongma over 4 years ago - 1 comment

#53 - Stopped h5py producing deprecation warning

Pull Request - State: open - Opened by EddyDavies over 4 years ago

#52 - Fix typo in TensorFlow README

Pull Request - State: closed - Opened by lukereichold over 4 years ago

#51 - About part segmentation.

Issue - State: open - Opened by a18700 over 4 years ago - 1 comment

#50 - reproduce the classification result with pytorch

Issue - State: closed - Opened by cslxiao over 4 years ago - 5 comments

#49 - Serious problem. About label smoothing in loss

Issue - State: closed - Opened by a18700 over 4 years ago - 2 comments

#48 - question about dataset

Issue - State: closed - Opened by NayeeC over 4 years ago - 1 comment

#47 - question about test for part segmentation

Issue - State: open - Opened by grxiao over 4 years ago

#46 - No such file or directory: 'pretrained/model.2048.t7'

Issue - State: closed - Opened by antao97 over 4 years ago - 3 comments

#45 - Where is the categorical vector?

Issue - State: closed - Opened by Steve-Tod over 4 years ago - 2 comments

#44 - Cuda memory

Issue - State: open - Opened by shersoni610 almost 5 years ago - 4 comments

#43 - Question about seg in S3DIS

Issue - State: open - Opened by MrCrazyCrab almost 5 years ago - 14 comments

#42 - Categoryless Training/Testing for part segmentation

Issue - State: closed - Opened by iKessho almost 5 years ago - 1 comment

#41 - Hi , can I train this model on my own dataset ?

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

#40 - About kernel_size

Issue - State: closed - Opened by mingzailao almost 5 years ago - 1 comment

#39 - The number of parameters of pytorch version

Issue - State: closed - Opened by Yachao-Zhang almost 5 years ago - 1 comment

#38 - InternalError (see above for traceback): Blas xGEMM launch failed

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

#37 - Why bias=None in conv2d?

Issue - State: closed - Opened by kxhit about 5 years ago - 1 comment

#36 - Undefined "weight_decay" in tensorflow/sem_seg/model.py

Issue - State: closed - Opened by MrJimm about 5 years ago - 3 comments

#35 - accuracy about classification

Issue - State: closed - Opened by grxiao about 5 years ago - 5 comments

#34 - Dose knn neighbours xj include the center point xi?

Issue - State: closed - Opened by wusuoweima about 5 years ago - 2 comments

#33 - How do you visualize the extracted features?

Issue - State: closed - Opened by longmalongma about 5 years ago - 1 comment

#32 - About spatial transform in pytorch implementation

Issue - State: closed - Opened by yfyang16 about 5 years ago - 1 comment

#31 - Training with my own data

Issue - State: closed - Opened by hongtaowu67 about 5 years ago - 1 comment

#30 - About Forward Time

Issue - State: closed - Opened by voidmomo about 5 years ago - 1 comment

#29 - visualization problem

Issue - State: closed - Opened by heng94 about 5 years ago - 1 comment

#28 - Questions about transform nets

Issue - State: closed - Opened by usherbob about 5 years ago - 1 comment

#27 - There exist no rotation augmentation in the preprocess of the pytorch implementation

Issue - State: closed - Opened by linhaojia13 about 5 years ago - 1 comment

#26 - segmentation experiments

Issue - State: closed - Opened by zed-2018 about 5 years ago - 2 comments

#25 - Why the test.py point_num=3000 ，However the train_multi_gpu.py point_num=2048 of part_seg ？

Issue - State: closed - Opened by shengrongjin about 5 years ago - 2 comments

#24 - PyTorch Implementation different with origin paper.

Issue - State: closed - Opened by XuyangBai about 5 years ago - 2 comments

#23 - The order of convolution and max operation.

Issue - State: closed - Opened by XuyangBai about 5 years ago - 2 comments

#22 - accuracy of ModelNet40

Issue - State: closed - Opened by rruixxu over 5 years ago - 2 comments

#21 - "The number of GPUs to use" in sem_seg with train.py

Issue - State: closed - Opened by suan0365006 over 5 years ago - 12 comments

#20 - An issue about the pairwise_distance

Issue - State: closed - Opened by sailor-z over 5 years ago - 2 comments

#19 - Questions about the Input point cloud data dimension of the EdgeConv

Issue - State: closed - Opened by limt15 over 5 years ago - 1 comment

#18 - Questions About the classification of ModelNet40

Issue - State: closed - Opened by wusuoweima over 5 years ago - 1 comment

#17 - Do we need multiple GPUs?

Issue - State: closed - Opened by usherbob over 5 years ago - 1 comment

#16 - Point cloud transform block different from paper

Issue - State: closed - Opened by patrick-llgc over 5 years ago - 1 comment

#15 - Visualization for part segmentation

Issue - State: closed - Opened by FrankCAN over 5 years ago - 2 comments

#14 - the difference between fixed knn graph and dynamic knn graph?

Issue - State: closed - Opened by Haiyan-Chris-Wang over 5 years ago - 5 comments

#13 - why do we need net = tf_util.conv2d(tf.concat([net1, net2, net3, net4], axis=-1), 1024, [1, 1], ?

Issue - State: closed - Opened by Haiyan-Chris-Wang over 5 years ago - 3 comments

#12 - accuracy issue

Issue - State: closed - Opened by FrankCAN over 5 years ago - 11 comments

#11 - The code is running super slow?

Issue - State: closed - Opened by Haiyan-Chris-Wang over 5 years ago - 6 comments

GitHub / WangYueFt/dgcnn issues and pull requests