From 4bfe95c9aa4a5f5f2d64352b53ed2d62983abb00 Mon Sep 17 00:00:00 2001 From: yiming Date: Wed, 23 Sep 2020 14:10:04 +0200 Subject: [PATCH 1/5] a quick test for fork --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index bf538ae..e5532e6 100644 --- a/README.md +++ b/README.md @@ -69,6 +69,7 @@ You can also visualise the point cloud at each time under each run and each meth ## Issues If you encounter issues in downloading the dataset or pre-trained models, please write to yiming.wang@iit.it, we will support you to download it. +Test fork! ## Citation If you find our work useful in your research, please consider citing: > @inproceedings{wang2020exhistcnn,\ From 91f10abed5b7b491a11e0de396e5ef03fd526ea9 Mon Sep 17 00:00:00 2001 From: yiming Date: Wed, 23 Sep 2020 14:34:34 +0200 Subject: [PATCH 2/5] revert the test --- README.md | 1 - 1 file changed, 1 deletion(-) diff --git a/README.md b/README.md index e5532e6..bf538ae 100644 --- a/README.md +++ b/README.md @@ -69,7 +69,6 @@ You can also visualise the point cloud at each time under each run and each meth ## Issues If you encounter issues in downloading the dataset or pre-trained models, please write to yiming.wang@iit.it, we will support you to download it. -Test fork! ## Citation If you find our work useful in your research, please consider citing: > @inproceedings{wang2020exhistcnn,\ From fe7f22099d193945548843081cb467fba420b32e Mon Sep 17 00:00:00 2001 From: yimingwang <19426699+yiming107@users.noreply.github.com> Date: Tue, 10 May 2022 15:49:04 +0200 Subject: [PATCH 3/5] updated the link to all files --- README.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index bf538ae..c888dc5 100644 --- a/README.md +++ b/README.md @@ -25,7 +25,7 @@ Activate the virtual environment, you are good to go :) ## Dataset -Download the dataset through the link. +Download the dataset through the link. The dataset contains the rendered rooms in SUNCG and Matterport3D which are used in this paper together with their ground-truth annotation json files. The renderings (depth+rgb) per room are generated with existing tools. For SUNCG, we use the [SUNCG toolbox](https://github.com/tinytangent/SUNCGtoolbox) while for real-world room scans from Matterport3D, we use [HabitatSim](https://github.com/facebookresearch/habitat-sim). You can find the detailed procedure of dataset generation in the video: @@ -38,7 +38,7 @@ alt="IMAGE ALT TEXT HERE" width="480" height="360" border="10" />
It is encouraged to put the dataset folder under the project, so that the script can be run without adapting paths. ### Dataset organisation -You can download the [csv files](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/Ec1AvAnEpehLgOWY2E3ZEJUB3gkNDVAUcxaGJkNzJDxj-Q?e=MLEy14) that organise the dataset for train, validation and test. +You can download the [csv files](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/Ec1AvAnEpehLgOWY2E3ZEJUB3gkNDVAUcxaGJkNzJDxj-Q?e=ACRZfB) that organise the dataset for train, validation and test. It is recommended to put the dataset organisation files in the data folder. Optionally, you can organise your own dataset split for training/validation/testing by tuning and running the scripts in *script/train/organise_dataset.py*. @@ -54,7 +54,7 @@ This should be done after you have the H5 files. ### Training You can run *script/train/trainExHistCNN.py* to train the ExHistCNN models. You can also find the corresponding scripts to train the model with only depth images and MLP classifiers. -Optionally you can download the pre-trained models from [here](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/EUBOjPb27VFOsNSuF7b8__EBoQ5WemMzOOxSJxdHyrnGAg?e=BfePqo) +Optionally you can download the pre-trained models from [here](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/EUBOjPb27VFOsNSuF7b8__EBoQ5WemMzOOxSJxdHyrnGAg?e=sx8InO) Please locate the pre-trained models within the checkpoint folder to avoid adapting path in the scripts. ### Evaluation @@ -62,7 +62,7 @@ You can evaluate the models by running: *script/train/evaluate_network.py*. The You can make the performance plot by running: *script/train/plot_result.py*. ## Visualise NBV results -You can download the metadata for the [results](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/EQ9TTdvz-f1MpV2OEjpJ6ksBqca_JQsJN2byhEvO7Vhutg?e=QiAXxe) for each NBV startegies reported in the paper. +You can download the metadata for the [results](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/EQ9TTdvz-f1MpV2OEjpJ6ksBw_icmt5N6Uq9nUg41RUDKQ?e=wcUOIE) for each NBV startegies reported in the paper. and run *script/result_analysis/analyse_result.py*, to obtain the figures as reported in the paper. You can also visualise the point cloud at each time under each run and each method by running *script/result_analysis/reconstruct_NBV.py*. From a63a2bd4db78837354944523d360af4f082ea9a5 Mon Sep 17 00:00:00 2001 From: yimingwang <19426699+yiming107@users.noreply.github.com> Date: Tue, 10 May 2022 15:51:28 +0200 Subject: [PATCH 4/5] update the links of all files --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index c888dc5..a62fdbe 100644 --- a/README.md +++ b/README.md @@ -25,7 +25,7 @@ Activate the virtual environment, you are good to go :) ## Dataset -Download the dataset through the link. +Download the dataset through the link. The dataset contains the rendered rooms in SUNCG and Matterport3D which are used in this paper together with their ground-truth annotation json files. The renderings (depth+rgb) per room are generated with existing tools. For SUNCG, we use the [SUNCG toolbox](https://github.com/tinytangent/SUNCGtoolbox) while for real-world room scans from Matterport3D, we use [HabitatSim](https://github.com/facebookresearch/habitat-sim). You can find the detailed procedure of dataset generation in the video: @@ -38,7 +38,7 @@ alt="IMAGE ALT TEXT HERE" width="480" height="360" border="10" />
It is encouraged to put the dataset folder under the project, so that the script can be run without adapting paths. ### Dataset organisation -You can download the [csv files](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/Ec1AvAnEpehLgOWY2E3ZEJUB3gkNDVAUcxaGJkNzJDxj-Q?e=ACRZfB) that organise the dataset for train, validation and test. +You can download the [csv files](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/Ec1AvAnEpehLgOWY2E3ZEJUB3gkNDVAUcxaGJkNzJDxj-Q?e=SK9rde) that organise the dataset for train, validation and test. It is recommended to put the dataset organisation files in the data folder. Optionally, you can organise your own dataset split for training/validation/testing by tuning and running the scripts in *script/train/organise_dataset.py*. From 1a0b61175fb17db1d1afccb2cf3e12c1c4ec7555 Mon Sep 17 00:00:00 2001 From: yimingwang <19426699+yiming107@users.noreply.github.com> Date: Tue, 10 May 2022 15:54:31 +0200 Subject: [PATCH 5/5] update the links --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index a62fdbe..0c4ae64 100644 --- a/README.md +++ b/README.md @@ -54,7 +54,7 @@ This should be done after you have the H5 files. ### Training You can run *script/train/trainExHistCNN.py* to train the ExHistCNN models. You can also find the corresponding scripts to train the model with only depth images and MLP classifiers. -Optionally you can download the pre-trained models from [here](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/EUBOjPb27VFOsNSuF7b8__EBoQ5WemMzOOxSJxdHyrnGAg?e=sx8InO) +Optionally you can download the pre-trained models from [here](https://istitutoitalianotecnologia-my.sharepoint.com/:u:/g/personal/yiming_wang_iit_it/EUBOjPb27VFOsNSuF7b8__EBoQ5WemMzOOxSJxdHyrnGAg?e=QKn4rz) Please locate the pre-trained models within the checkpoint folder to avoid adapting path in the scripts. ### Evaluation