Yongsen Mao
commited on
Commit
•
4960dfe
1
Parent(s):
ab48b74
Add VP Hough Transform Gaussian Sphere models
Browse files- README.md +2 -0
- dsine/LICENSE +230 -0
- dsine/README.md +16 -0
- vp_houghtransform_gaussiansphere/LICENSE +21 -0
- vp_houghtransform_gaussiansphere/NYU/checkpoint_latest.pth.tar +3 -0
- vp_houghtransform_gaussiansphere/NYU/config.yaml +28 -0
- vp_houghtransform_gaussiansphere/README.md +16 -0
- vp_houghtransform_gaussiansphere/SU3/checkpoint_latest.pth.tar +3 -0
- vp_houghtransform_gaussiansphere/SU3/config.yaml +28 -0
- vp_houghtransform_gaussiansphere/SU3_multi_scale/checkpoint_latest.pth.tar +3 -0
- vp_houghtransform_gaussiansphere/ScanNet/checkpoint_latest.pth.tar +3 -0
- vp_houghtransform_gaussiansphere/ScanNet/config.yaml +28 -0
README.md
CHANGED
@@ -6,6 +6,8 @@ license: mit
|
|
6 |
|
7 |
PySceneKit is an open-source Python library designed for common scene processing and visualization tasks. Whether you're working with 2D or 3D scenes, PySceneKit provides a comprehensive toolkit to help you manipulate, analyze, and visualize your data with ease.
|
8 |
|
|
|
|
|
9 |
## Introduction
|
10 |
|
11 |
Welcome to PySceneKit! This project is fueled by my passion for scene understanding, particularly in indoor environments. Frustrated by the lack of user-friendly tools for processing images and 3D indoor scenes, I often found myself rewriting code for different datasets.
|
|
|
6 |
|
7 |
PySceneKit is an open-source Python library designed for common scene processing and visualization tasks. Whether you're working with 2D or 3D scenes, PySceneKit provides a comprehensive toolkit to help you manipulate, analyze, and visualize your data with ease.
|
8 |
|
9 |
+
GitHub: https://github.com/SamMaoYS/pyscenekit
|
10 |
+
|
11 |
## Introduction
|
12 |
|
13 |
Welcome to PySceneKit! This project is fueled by my passion for scene understanding, particularly in indoor environments. Frustrated by the lack of user-friendly tools for processing images and 3D indoor scenes, I often found myself rewriting code for different datasets.
|
dsine/LICENSE
ADDED
@@ -0,0 +1,230 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
DSINE SOFTWARE
|
2 |
+
|
3 |
+
LICENCE AGREEMENT
|
4 |
+
|
5 |
+
WE (Imperial College of Science, Technology and Medicine, (“Imperial College
|
6 |
+
London”)) ARE WILLING TO LICENSE THIS SOFTWARE TO YOU (a licensee “You”) ONLY
|
7 |
+
ON THE CONDITION THAT YOU ACCEPT ALL OF THE TERMS CONTAINED IN THE FOLLOWING
|
8 |
+
AGREEMENT. PLEASE READ THE AGREEMENT CAREFULLY BEFORE DOWNLOADING THE SOFTWARE.
|
9 |
+
BY EXERCISING THE OPTION TO DOWNLOAD THE SOFTWARE YOU AGREE TO BE BOUND BY THE
|
10 |
+
TERMS OF THE AGREEMENT.
|
11 |
+
|
12 |
+
SOFTWARE LICENCE AGREEMENT (EXCLUDING BSD COMPONENTS)
|
13 |
+
|
14 |
+
1. This Agreement pertains to a worldwide, non-exclusive, temporary, fully
|
15 |
+
paid-up, royalty free, non-transferable, non-sub- licensable licence (the
|
16 |
+
“Licence”) to use the elastic fusion source code, including any modification,
|
17 |
+
part or derivative (the “Software”).
|
18 |
+
|
19 |
+
Ownership and Licence. Your rights to use and download the Software onto your
|
20 |
+
computer, and all other copies that You are authorised to make, are specified
|
21 |
+
in this Agreement. However, we (or our licensors) retain all rights, including
|
22 |
+
but not limited to all copyright and other intellectual property rights
|
23 |
+
anywhere in the world, in the Software not expressly granted to You in this
|
24 |
+
Agreement.
|
25 |
+
|
26 |
+
2. Permitted use of the Licence:
|
27 |
+
|
28 |
+
(a) You may download and install the Software onto one computer or server for
|
29 |
+
use in accordance with Clause 2(b) of this Agreement provided that You ensure
|
30 |
+
that the Software is not accessible by other users unless they have themselves
|
31 |
+
accepted the terms of this licence agreement.
|
32 |
+
|
33 |
+
(b) You may use the Software solely for non-commercial, internal or academic
|
34 |
+
research purposes and only in accordance with the terms of this Agreement. You
|
35 |
+
may not use the Software for commercial purposes, including but not limited to
|
36 |
+
(1) integration of all or part of the source code or the Software into a
|
37 |
+
product for sale or licence by or on behalf of You to third parties or (2) use
|
38 |
+
of the Software or any derivative of it for research to develop software
|
39 |
+
products for sale or licence to a third party or (3) use of the Software or any
|
40 |
+
derivative of it for research to develop non-software products for sale or
|
41 |
+
licence to a third party, or (4) use of the Software to provide any service to
|
42 |
+
an external organisation for which payment is received.
|
43 |
+
|
44 |
+
Should You wish to use the Software for commercial purposes, You shall
|
45 |
+
email researchcontracts.engineering@imperial.ac.uk .
|
46 |
+
|
47 |
+
(c) Right to Copy. You may copy the Software for back-up and archival purposes,
|
48 |
+
provided that each copy is kept in your possession and provided You reproduce
|
49 |
+
our copyright notice (set out in Schedule 1) on each copy.
|
50 |
+
|
51 |
+
(d) Transfer and sub-licensing. You may not rent, lend, or lease the Software
|
52 |
+
and You may not transmit, transfer or sub-license this licence to use the
|
53 |
+
Software or any of your rights or obligations under this Agreement to another
|
54 |
+
party.
|
55 |
+
|
56 |
+
(e) Identity of Licensee. The licence granted herein is personal to You. You
|
57 |
+
shall not permit any third party to access, modify or otherwise use the
|
58 |
+
Software nor shall You access modify or otherwise use the Software on behalf of
|
59 |
+
any third party. If You wish to obtain a licence for mutiple users or a site
|
60 |
+
licence for the Software please contact us
|
61 |
+
at researchcontracts.engineering@imperial.ac.uk .
|
62 |
+
|
63 |
+
(f) Publications and presentations. You may make public, results or data
|
64 |
+
obtained from, dependent on or arising from research carried out using the
|
65 |
+
Software, provided that any such presentation or publication identifies the
|
66 |
+
Software as the source of the results or the data, including the Copyright
|
67 |
+
Notice given in each element of the Software, and stating that the Software has
|
68 |
+
been made available for use by You under licence from Imperial College London
|
69 |
+
and You provide a copy of any such publication to Imperial College London.
|
70 |
+
|
71 |
+
3. Prohibited Uses. You may not, without written permission from us
|
72 |
+
at researchcontracts.engineering@imperial.ac.uk :
|
73 |
+
|
74 |
+
(a) Use, copy, modify, merge, or transfer copies of the Software or any
|
75 |
+
documentation provided by us which relates to the Software except as provided
|
76 |
+
in this Agreement;
|
77 |
+
|
78 |
+
(b) Use any back-up or archival copies of the Software (or allow anyone else to
|
79 |
+
use such copies) for any purpose other than to replace the original copy in the
|
80 |
+
event it is destroyed or becomes defective; or
|
81 |
+
|
82 |
+
(c) Disassemble, decompile or "unlock", reverse translate, or in any manner
|
83 |
+
decode the Software for any reason.
|
84 |
+
|
85 |
+
4. Warranty Disclaimer
|
86 |
+
|
87 |
+
(a) Disclaimer. The Software has been developed for research purposes only. You
|
88 |
+
acknowledge that we are providing the Software to You under this licence
|
89 |
+
agreement free of charge and on condition that the disclaimer set out below
|
90 |
+
shall apply. We do not represent or warrant that the Software as to: (i) the
|
91 |
+
quality, accuracy or reliability of the Software; (ii) the suitability of the
|
92 |
+
Software for any particular use or for use under any specific conditions; and
|
93 |
+
(iii) whether use of the Software will infringe third-party rights.
|
94 |
+
|
95 |
+
You acknowledge that You have reviewed and evaluated the Software to determine
|
96 |
+
that it meets your needs and that You assume all responsibility and liability
|
97 |
+
for determining the suitability of the Software as fit for your particular
|
98 |
+
purposes and requirements. Subject to Clause 4(b), we exclude and expressly
|
99 |
+
disclaim all express and implied representations, warranties, conditions and
|
100 |
+
terms not stated herein (including the implied conditions or warranties of
|
101 |
+
satisfactory quality, merchantable quality, merchantability and fitness for
|
102 |
+
purpose).
|
103 |
+
|
104 |
+
(b) Savings. Some jurisdictions may imply warranties, conditions or terms or
|
105 |
+
impose obligations upon us which cannot, in whole or in part, be excluded,
|
106 |
+
restricted or modified or otherwise do not allow the exclusion of implied
|
107 |
+
warranties, conditions or terms, in which case the above warranty disclaimer
|
108 |
+
and exclusion will only apply to You to the extent permitted in the relevant
|
109 |
+
jurisdiction and does not in any event exclude any implied warranties,
|
110 |
+
conditions or terms which may not under applicable law be excluded.
|
111 |
+
|
112 |
+
(c) Imperial College London disclaims all responsibility for the use which is
|
113 |
+
made of the Software and any liability for the outcomes arising from using the
|
114 |
+
Software.
|
115 |
+
|
116 |
+
5. Limitation of Liability
|
117 |
+
|
118 |
+
(a) You acknowledge that we are providing the Software to You under this
|
119 |
+
licence agreement free of charge and on condition that the limitation of
|
120 |
+
liability set out below shall apply. Accordingly, subject to Clause 5(b), we
|
121 |
+
exclude all liability whether in contract, tort, negligence or otherwise, in
|
122 |
+
respect of the Software and/or any related documentation provided to You by us
|
123 |
+
including, but not limited to, liability for loss or corruption of data, loss
|
124 |
+
of contracts, loss of income, loss of profits, loss of cover and any
|
125 |
+
consequential or indirect loss or damage of any kind arising out of or in
|
126 |
+
connection with this licence agreement, however caused. This exclusion shall
|
127 |
+
apply even if we have been advised of the possibility of such loss or damage.
|
128 |
+
|
129 |
+
(b) You agree to indemnify Imperial College London and hold it harmless from
|
130 |
+
and against any and all claims, damages and liabilities asserted by third
|
131 |
+
parties (including claims for negligence) which arise directly or indirectly
|
132 |
+
from the use of the Software or any derivative of it or the sale of any
|
133 |
+
products based on the Software. You undertake to make no liability claim
|
134 |
+
against any employee, student, agent or appointee of Imperial College London,
|
135 |
+
in connection with this Licence or the Software.
|
136 |
+
|
137 |
+
(c) Nothing in this Agreement shall have the effect of excluding or limiting
|
138 |
+
our statutory liability.
|
139 |
+
|
140 |
+
(d) Some jurisdictions do not allow these limitations or exclusions either
|
141 |
+
wholly or in part, and, to that extent, they may not apply to you. Nothing in
|
142 |
+
this licence agreement will affect your statutory rights or other relevant
|
143 |
+
statutory provisions which cannot be excluded, restricted or modified, and its
|
144 |
+
terms and conditions must be read and construed subject to any such statutory
|
145 |
+
rights and/or provisions.
|
146 |
+
|
147 |
+
6. Confidentiality. You agree not to disclose any confidential information
|
148 |
+
provided to You by us pursuant to this Agreement to any third party without our
|
149 |
+
prior written consent. The obligations in this Clause 6 shall survive the
|
150 |
+
termination of this Agreement for any reason.
|
151 |
+
|
152 |
+
7. Termination.
|
153 |
+
|
154 |
+
(a) We may terminate this licence agreement and your right to use the Software
|
155 |
+
at any time with immediate effect upon written notice to You.
|
156 |
+
|
157 |
+
(b) This licence agreement and your right to use the Software automatically
|
158 |
+
terminate if You:
|
159 |
+
|
160 |
+
(i) fail to comply with any provisions of this Agreement; or
|
161 |
+
|
162 |
+
(ii) destroy the copies of the Software in your possession, or voluntarily
|
163 |
+
return the Software to us.
|
164 |
+
|
165 |
+
(c) Upon termination You will destroy all copies of the Software.
|
166 |
+
|
167 |
+
(d) Otherwise, the restrictions on your rights to use the Software will expire
|
168 |
+
10 (ten) years after first use of the Software under this licence agreement.
|
169 |
+
|
170 |
+
8. Miscellaneous Provisions.
|
171 |
+
|
172 |
+
(a) This Agreement will be governed by and construed in accordance with the
|
173 |
+
substantive laws of England and Wales whose courts shall have exclusive
|
174 |
+
jurisdiction over all disputes which may arise between us.
|
175 |
+
|
176 |
+
(b) This is the entire agreement between us relating to the Software, and
|
177 |
+
supersedes any prior purchase order, communications, advertising or
|
178 |
+
representations concerning the Software.
|
179 |
+
|
180 |
+
(c) No change or modification of this Agreement will be valid unless it is in
|
181 |
+
writing, and is signed by us.
|
182 |
+
|
183 |
+
(d) The unenforceability or invalidity of any part of this Agreement will not
|
184 |
+
affect the enforceability or validity of the remaining parts.
|
185 |
+
|
186 |
+
BSD Elements of the Software
|
187 |
+
|
188 |
+
For BSD elements of the Software, the following terms shall apply:
|
189 |
+
Copyright as indicated in the header of the individual element of the Software.
|
190 |
+
All rights reserved.
|
191 |
+
|
192 |
+
Redistribution and use in source and binary forms, with or without
|
193 |
+
modification, are permitted provided that the following conditions are met:
|
194 |
+
|
195 |
+
1. Redistributions of source code must retain the above copyright notice, this
|
196 |
+
list of conditions and the following disclaimer.
|
197 |
+
|
198 |
+
2. Redistributions in binary form must reproduce the above copyright notice,
|
199 |
+
this list of conditions and the following disclaimer in the documentation
|
200 |
+
and/or other materials provided with the distribution.
|
201 |
+
|
202 |
+
3. Neither the name of the copyright holder nor the names of its contributors
|
203 |
+
may be used to endorse or promote products derived from this software without
|
204 |
+
specific prior written permission.
|
205 |
+
|
206 |
+
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
|
207 |
+
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
|
208 |
+
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
209 |
+
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
|
210 |
+
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
211 |
+
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
|
212 |
+
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
213 |
+
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
214 |
+
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
215 |
+
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
216 |
+
|
217 |
+
SCHEDULE 1
|
218 |
+
|
219 |
+
The Software
|
220 |
+
|
221 |
+
DSINE is a framework for estimating surface normals from a single image. It is based on the techniques described in the following publication:
|
222 |
+
|
223 |
+
• Gwangbin Bae, Andrew J. Davison. Rethinking Inductive Biases for Surface Normal Estimation. CVPR, 2024
|
224 |
+
_________________________
|
225 |
+
|
226 |
+
Acknowledgments
|
227 |
+
|
228 |
+
If you use the software, you should reference the following paper in any publication:
|
229 |
+
|
230 |
+
• Gwangbin Bae, Andrew J. Davison. Rethinking Inductive Biases for Surface Normal Estimation. CVPR, 2024
|
dsine/README.md
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# DSINE
|
2 |
+
|
3 |
+
Rethinking Inductive Biases for Surface Normal Estimation
|
4 |
+
|
5 |
+
Authors: Gwangbin Bae, Andrew J. Davison
|
6 |
+
|
7 |
+
https://github.com/baegwangbin/DSINE
|
8 |
+
|
9 |
+
```bibtex
|
10 |
+
@inproceedings{bae2024dsine,
|
11 |
+
title = {Rethinking Inductive Biases for Surface Normal Estimation},
|
12 |
+
author = {Gwangbin Bae and Andrew J. Davison},
|
13 |
+
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
|
14 |
+
year = {2024}
|
15 |
+
}
|
16 |
+
```
|
vp_houghtransform_gaussiansphere/LICENSE
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
MIT License
|
2 |
+
|
3 |
+
Copyright (c) 2022 CVisioner
|
4 |
+
|
5 |
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
6 |
+
of this software and associated documentation files (the "Software"), to deal
|
7 |
+
in the Software without restriction, including without limitation the rights
|
8 |
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
9 |
+
copies of the Software, and to permit persons to whom the Software is
|
10 |
+
furnished to do so, subject to the following conditions:
|
11 |
+
|
12 |
+
The above copyright notice and this permission notice shall be included in all
|
13 |
+
copies or substantial portions of the Software.
|
14 |
+
|
15 |
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
16 |
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
17 |
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
18 |
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
19 |
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
20 |
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
21 |
+
SOFTWARE.
|
vp_houghtransform_gaussiansphere/NYU/checkpoint_latest.pth.tar
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:72a77c460fc35529a1f6a3f1abfeb66a96cb8e5a467f36241964ed5f50a2e90c
|
3 |
+
size 117820242
|
vp_houghtransform_gaussiansphere/NYU/config.yaml
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
io:
|
2 |
+
datadir: dataset/nyu_vp/processed_data_aug
|
3 |
+
dataset: NYU
|
4 |
+
focal_length: 1.0
|
5 |
+
logdir: logs/
|
6 |
+
num_neighbors: 20
|
7 |
+
num_nodes: 32768
|
8 |
+
num_vpts: null
|
9 |
+
num_workers: 4
|
10 |
+
percentage: 1
|
11 |
+
resume_from: logs/211103-130207-nyu
|
12 |
+
tensorboard_port: 0
|
13 |
+
validation_debug: 120
|
14 |
+
model:
|
15 |
+
backbone: stacked_hourglass
|
16 |
+
batch_size: 3
|
17 |
+
depth: 4
|
18 |
+
lneg: 1.0
|
19 |
+
lpos: 1.0
|
20 |
+
num_blocks: 1
|
21 |
+
num_stacks: 1
|
22 |
+
optim:
|
23 |
+
amsgrad: true
|
24 |
+
lr: 0.0004
|
25 |
+
lr_decay_epoch: 24
|
26 |
+
max_epoch: 36
|
27 |
+
name: Adam
|
28 |
+
weight_decay: 1.0e-05
|
vp_houghtransform_gaussiansphere/README.md
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# VP Hough Transform Gaussian Sphere
|
2 |
+
|
3 |
+
Deep vanishing point detection: Geometric priors make dataset variations vanish
|
4 |
+
|
5 |
+
Authors: Yancong Lin, Ruben Wiersma, Silvia L. Pintea, Klaus Hildebrandt, Elmar Eisemann, Jan C. van Gemert
|
6 |
+
|
7 |
+
Original repo: https://github.com/yanconglin/VanishingPoint_HoughTransform_GaussianSphere
|
8 |
+
|
9 |
+
```bibtex
|
10 |
+
@article{lin2022vpd,
|
11 |
+
title={Deep vanishing point detection: Geometric priors make dataset variations vanish},
|
12 |
+
author={Lin, Yancong and Wiersma, Ruben and and Pintea, Silvia L and Hildebrandt, Klaus and Eisemann, Elmar and van Gemert, Jan C},
|
13 |
+
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
|
14 |
+
year={2022}
|
15 |
+
}
|
16 |
+
```
|
vp_houghtransform_gaussiansphere/SU3/checkpoint_latest.pth.tar
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9964f36c23c64c1f357112e5cfb492c4e78537e49be5f6c1b7ce8e3ef52c642f
|
3 |
+
size 115723090
|
vp_houghtransform_gaussiansphere/SU3/config.yaml
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
io:
|
2 |
+
datadir: dataset/su3
|
3 |
+
dataset: Wireframe
|
4 |
+
focal_length: 2.1875
|
5 |
+
logdir: logs/
|
6 |
+
num_neighbors: 16
|
7 |
+
num_nodes: 32768
|
8 |
+
num_vpts: 3
|
9 |
+
num_workers: 2
|
10 |
+
percentage: 1.0
|
11 |
+
resume_from: logs/220228-092845-su3
|
12 |
+
tensorboard_port: 0
|
13 |
+
validation_debug: 120
|
14 |
+
model:
|
15 |
+
backbone: stacked_hourglass
|
16 |
+
batch_size: 4
|
17 |
+
depth: 4
|
18 |
+
lneg: 1.0
|
19 |
+
lpos: 1.0
|
20 |
+
num_blocks: 1
|
21 |
+
num_stacks: 1
|
22 |
+
optim:
|
23 |
+
amsgrad: true
|
24 |
+
lr: 0.0004
|
25 |
+
lr_decay_epoch: 24
|
26 |
+
max_epoch: 36
|
27 |
+
name: Adam
|
28 |
+
weight_decay: 1.0e-05
|
vp_houghtransform_gaussiansphere/SU3_multi_scale/checkpoint_latest.pth.tar
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b3fa02941d55f1d2653b555adf55872776fec90e7c647d0db30347ae3623e972
|
3 |
+
size 84934045
|
vp_houghtransform_gaussiansphere/ScanNet/checkpoint_latest.pth.tar
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:70c53b70963578f7b7548f6a12391e12918a92a990f9d5d488b959597bb53d7a
|
3 |
+
size 113260242
|
vp_houghtransform_gaussiansphere/ScanNet/config.yaml
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
io:
|
2 |
+
datadir: dataset/scannet-vp
|
3 |
+
dataset: ScanNet
|
4 |
+
focal_length: 2.408333333333333
|
5 |
+
logdir: logs/
|
6 |
+
num_neighbors: 16
|
7 |
+
num_nodes: 16384
|
8 |
+
num_vpts: 3
|
9 |
+
num_workers: 4
|
10 |
+
percentage: 1.0
|
11 |
+
resume_from: logs/211031-020102-scannet
|
12 |
+
tensorboard_port: 0
|
13 |
+
validation_debug: 240
|
14 |
+
model:
|
15 |
+
backbone: stacked_hourglass
|
16 |
+
batch_size: 4
|
17 |
+
depth: 4
|
18 |
+
lneg: 1.0
|
19 |
+
lpos: 1.0
|
20 |
+
num_blocks: 1
|
21 |
+
num_stacks: 1
|
22 |
+
optim:
|
23 |
+
amsgrad: true
|
24 |
+
lr: 0.0004
|
25 |
+
lr_decay_epoch: 4
|
26 |
+
max_epoch: 10
|
27 |
+
name: Adam
|
28 |
+
weight_decay: 1.0e-05
|