Skip to content

Commit f63aadf

Browse files
author
LoickCh
committed
init
0 parents  commit f63aadf

94 files changed

Lines changed: 7689 additions & 0 deletions

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎.gitignore‎

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
data/
2+
__pycache__/
3+
*.nfs*
4+
output/
5+
.vscode

‎LICENSE‎

Lines changed: 202 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,202 @@
1+
2+
Apache License
3+
Version 2.0, January 2004
4+
http://www.apache.org/licenses/
5+
6+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7+
8+
1. Definitions.
9+
10+
"License" shall mean the terms and conditions for use, reproduction,
11+
and distribution as defined by Sections 1 through 9 of this document.
12+
13+
"Licensor" shall mean the copyright owner or entity authorized by
14+
the copyright owner that is granting the License.
15+
16+
"Legal Entity" shall mean the union of the acting entity and all
17+
other entities that control, are controlled by, or are under common
18+
control with that entity. For the purposes of this definition,
19+
"control" means (i) the power, direct or indirect, to cause the
20+
direction or management of such entity, whether by contract or
21+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
22+
outstanding shares, or (iii) beneficial ownership of such entity.
23+
24+
"You" (or "Your") shall mean an individual or Legal Entity
25+
exercising permissions granted by this License.
26+
27+
"Source" form shall mean the preferred form for making modifications,
28+
including but not limited to software source code, documentation
29+
source, and configuration files.
30+
31+
"Object" form shall mean any form resulting from mechanical
32+
transformation or translation of a Source form, including but
33+
not limited to compiled object code, generated documentation,
34+
and conversions to other media types.
35+
36+
"Work" shall mean the work of authorship, whether in Source or
37+
Object form, made available under the License, as indicated by a
38+
copyright notice that is included in or attached to the work
39+
(an example is provided in the Appendix below).
40+
41+
"Derivative Works" shall mean any work, whether in Source or Object
42+
form, that is based on (or derived from) the Work and for which the
43+
editorial revisions, annotations, elaborations, or other modifications
44+
represent, as a whole, an original work of authorship. For the purposes
45+
of this License, Derivative Works shall not include works that remain
46+
separable from, or merely link (or bind by name) to the interfaces of,
47+
the Work and Derivative Works thereof.
48+
49+
"Contribution" shall mean any work of authorship, including
50+
the original version of the Work and any modifications or additions
51+
to that Work or Derivative Works thereof, that is intentionally
52+
submitted to Licensor for inclusion in the Work by the copyright owner
53+
or by an individual or Legal Entity authorized to submit on behalf of
54+
the copyright owner. For the purposes of this definition, "submitted"
55+
means any form of electronic, verbal, or written communication sent
56+
to the Licensor or its representatives, including but not limited to
57+
communication on electronic mailing lists, source code control systems,
58+
and issue tracking systems that are managed by, or on behalf of, the
59+
Licensor for the purpose of discussing and improving the Work, but
60+
excluding communication that is conspicuously marked or otherwise
61+
designated in writing by the copyright owner as "Not a Contribution."
62+
63+
"Contributor" shall mean Licensor and any individual or Legal Entity
64+
on behalf of whom a Contribution has been received by Licensor and
65+
subsequently incorporated within the Work.
66+
67+
2. Grant of Copyright License. Subject to the terms and conditions of
68+
this License, each Contributor hereby grants to You a perpetual,
69+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70+
copyright license to reproduce, prepare Derivative Works of,
71+
publicly display, publicly perform, sublicense, and distribute the
72+
Work and such Derivative Works in Source or Object form.
73+
74+
3. Grant of Patent License. Subject to the terms and conditions of
75+
this License, each Contributor hereby grants to You a perpetual,
76+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77+
(except as stated in this section) patent license to make, have made,
78+
use, offer to sell, sell, import, and otherwise transfer the Work,
79+
where such license applies only to those patent claims licensable
80+
by such Contributor that are necessarily infringed by their
81+
Contribution(s) alone or by combination of their Contribution(s)
82+
with the Work to which such Contribution(s) was submitted. If You
83+
institute patent litigation against any entity (including a
84+
cross-claim or counterclaim in a lawsuit) alleging that the Work
85+
or a Contribution incorporated within the Work constitutes direct
86+
or contributory patent infringement, then any patent licenses
87+
granted to You under this License for that Work shall terminate
88+
as of the date such litigation is filed.
89+
90+
4. Redistribution. You may reproduce and distribute copies of the
91+
Work or Derivative Works thereof in any medium, with or without
92+
modifications, and in Source or Object form, provided that You
93+
meet the following conditions:
94+
95+
(a) You must give any other recipients of the Work or
96+
Derivative Works a copy of this License; and
97+
98+
(b) You must cause any modified files to carry prominent notices
99+
stating that You changed the files; and
100+
101+
(c) You must retain, in the Source form of any Derivative Works
102+
that You distribute, all copyright, patent, trademark, and
103+
attribution notices from the Source form of the Work,
104+
excluding those notices that do not pertain to any part of
105+
the Derivative Works; and
106+
107+
(d) If the Work includes a "NOTICE" text file as part of its
108+
distribution, then any Derivative Works that You distribute must
109+
include a readable copy of the attribution notices contained
110+
within such NOTICE file, excluding those notices that do not
111+
pertain to any part of the Derivative Works, in at least one
112+
of the following places: within a NOTICE text file distributed
113+
as part of the Derivative Works; within the Source form or
114+
documentation, if provided along with the Derivative Works; or,
115+
within a display generated by the Derivative Works, if and
116+
wherever such third-party notices normally appear. The contents
117+
of the NOTICE file are for informational purposes only and
118+
do not modify the License. You may add Your own attribution
119+
notices within Derivative Works that You distribute, alongside
120+
or as an addendum to the NOTICE text from the Work, provided
121+
that such additional attribution notices cannot be construed
122+
as modifying the License.
123+
124+
You may add Your own copyright statement to Your modifications and
125+
may provide additional or different license terms and conditions
126+
for use, reproduction, or distribution of Your modifications, or
127+
for any such Derivative Works as a whole, provided Your use,
128+
reproduction, and distribution of the Work otherwise complies with
129+
the conditions stated in this License.
130+
131+
5. Submission of Contributions. Unless You explicitly state otherwise,
132+
any Contribution intentionally submitted for inclusion in the Work
133+
by You to the Licensor shall be under the terms and conditions of
134+
this License, without any additional terms or conditions.
135+
Notwithstanding the above, nothing herein shall supersede or modify
136+
the terms of any separate license agreement you may have executed
137+
with Licensor regarding such Contributions.
138+
139+
6. Trademarks. This License does not grant permission to use the trade
140+
names, trademarks, service marks, or product names of the Licensor,
141+
except as required for reasonable and customary use in describing the
142+
origin of the Work and reproducing the content of the NOTICE file.
143+
144+
7. Disclaimer of Warranty. Unless required by applicable law or
145+
agreed to in writing, Licensor provides the Work (and each
146+
Contributor provides its Contributions) on an "AS IS" BASIS,
147+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148+
implied, including, without limitation, any warranties or conditions
149+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150+
PARTICULAR PURPOSE. You are solely responsible for determining the
151+
appropriateness of using or redistributing the Work and assume any
152+
risks associated with Your exercise of permissions under this License.
153+
154+
8. Limitation of Liability. In no event and under no legal theory,
155+
whether in tort (including negligence), contract, or otherwise,
156+
unless required by applicable law (such as deliberate and grossly
157+
negligent acts) or agreed to in writing, shall any Contributor be
158+
liable to You for damages, including any direct, indirect, special,
159+
incidental, or consequential damages of any character arising as a
160+
result of this License or out of the use or inability to use the
161+
Work (including but not limited to damages for loss of goodwill,
162+
work stoppage, computer failure or malfunction, or any and all
163+
other commercial damages or losses), even if such Contributor
164+
has been advised of the possibility of such damages.
165+
166+
9. Accepting Warranty or Additional Liability. While redistributing
167+
the Work or Derivative Works thereof, You may choose to offer,
168+
and charge a fee for, acceptance of support, warranty, indemnity,
169+
or other liability obligations and/or rights consistent with this
170+
License. However, in accepting such obligations, You may act only
171+
on Your own behalf and on Your sole responsibility, not on behalf
172+
of any other Contributor, and only if You agree to indemnify,
173+
defend, and hold each Contributor harmless for any liability
174+
incurred by, or claims asserted against, such Contributor by reason
175+
of your accepting any such warranty or additional liability.
176+
177+
END OF TERMS AND CONDITIONS
178+
179+
APPENDIX: How to apply the Apache License to your work.
180+
181+
To apply the Apache License to your work, attach the following
182+
boilerplate notice, with the fields enclosed by brackets "[]"
183+
replaced with your own identifying information. (Don't include
184+
the brackets!) The text should be enclosed in the appropriate
185+
comment syntax for the file format. We also recommend that a
186+
file or class name and description of purpose be included on the
187+
same "printed page" as the copyright notice for easier
188+
identification within third-party archives.
189+
190+
Copyright [yyyy] [name of copyright owner]
191+
192+
Licensed under the Apache License, Version 2.0 (the "License");
193+
you may not use this file except in compliance with the License.
194+
You may obtain a copy of the License at
195+
196+
http://www.apache.org/licenses/LICENSE-2.0
197+
198+
Unless required by applicable law or agreed to in writing, software
199+
distributed under the License is distributed on an "AS IS" BASIS,
200+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201+
See the License for the specific language governing permissions and
202+
limitations under the License.

‎README.md‎

Lines changed: 180 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,180 @@
1+
# Official Implementation of *NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering.*
2+
3+
> [**NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering.**]<br>
4+
> [Loick Chambon](https://loickch.github.io/), [Paul Couairon](https://pcouairon.github.io/), [Eloi Zablocki](https://scholar.google.fr/citations?user=dOkbUmEAAAAJ&hl=fr), [Alexandre Boulch](https://boulch.eu/), [Nicolas Thome](https://thome.isir.upmc.fr/), [Matthieu Cord](https://cord.isir.upmc.fr/).<br> Valeo.ai, Sorbonne University, CNRS.<br>
5+
6+
7+
<table>
8+
<tr>
9+
<td align="center" width="50%">
10+
<img src='./asset/teasing.gif' width="90%">
11+
</td>
12+
</tr>
13+
</table>
14+
15+
## 🎯 TL;DR
16+
17+
**Three simple steps:**
18+
1. **Select any Vision Foundation Model** ([DINOv3](https://github.com/facebookresearch/dinov3), [DINOv2](https://github.com/facebookresearch/dinov2), [RADIO](https://github.com/NVlabs/RADIO), [FRANCA](https://github.com/valeoai/Franca), [PE-CORE](https://github.com/facebookresearch/perception_models), [CLIP](https://github.com/openai/CLIP), [SAM](https://github.com/facebookresearch/segment-anything), etc.)
19+
2. **Choose your target resolution** (up to 2K)
20+
3. **Upsample features with NAF** — zero-shot, no retraining needed
21+
22+
**Why it works:** NAF combines classical filtering theory with modern attention mechanisms, learning adaptive kernels through Fourier space transformations.
23+
24+
## ⚡ News & Updates
25+
- [ ] Release trained checkpoints for **NAF++**.
26+
- [x] **[2025-11-25]** NAF has been uploaded on arXiv.
27+
- [x] **[2025-11-24]** NAF code has been publicly released.
28+
29+
# 📜 Abstract
30+
31+
Vision Foundation Models produce **downsampled spatial features**, which are challenging for pixel-level tasks.
32+
33+
❌ Traditional upsampling methods:
34+
* **Classical filters** – fast, generic, but fixed (bilinear, bicubic, joint bilateral, guided)
35+
* **Learnable VFM-specific upsamplers** – accurate, but need retraining (FeatUp, LiFT, JAFAR, LoftUp)
36+
37+
✅ **NAF (Neighborhood Attention Filtering)**:
38+
* Learns **adaptive spatial-and-content weights** using Cross-Scale Neighborhood Attention + RoPE
39+
* Works **zero-shot** for any VFM
40+
* Outperforms existing upsamplers on **multiple downstream tasks**
41+
* Efficient: scales up to 2K features, ~18 FPS for intermediate resolutions
42+
* Also effective for **image restoration**
43+
44+
45+
# 🥇 Results
46+
47+
NAF achieves **state-of-the-art performance** on:
48+
49+
- Semantic Segmentation
50+
- Depth Estimation
51+
- Open-Vocabulary Tasks
52+
- Video Segmentation Propagation
53+
54+
Tested with **multiple VFMs** (T-S-B-L-G-7B) and **datasets** (COCO, VOC, ADE20K, Cityscapes, KITTI-360, NYUv2)
55+
56+
57+
## ✨ Downstream Tasks
58+
59+
<table>
60+
<tr>
61+
<td align="center">
62+
<img src="asset/results.png" style="width: 90%;">
63+
</td>
64+
</tr>
65+
66+
<tr>
67+
<td align="center">
68+
<em>NAF achieves state-of-the-art performance on various benchmarks: Semantic Segmentation, Depth Estimation, Video Label Propagation, Open-Vocabulary beating previous VFM-specific and VFM-agnostic methods while being more efficient.</em>
69+
</td>
70+
</tr>
71+
</table>
72+
73+
## 🔢 Filtering and Fourier Analysis
74+
75+
We found that NAF learns the Inverse Discrete Fourier Transform (IDFT) of the upsampling aggregation kernel, providing insights into the mechanism behind its results:
76+
77+
**NAF Filtering equation:**
78+
79+
$$
80+
\mathbf{F}^{\mathrm{HR}}_{p} = \frac{1}{Z(p)} \sum_{q \in \mathcal{N}(p)} \exp\left(\frac{\langle Q_p, K_q \rangle}{\sqrt{d}} \right) \mathbf{F}^\mathrm{LR}_{q}
81+
$$
82+
83+
**NAF Fourier interpretation:**
84+
85+
$$
86+
S(x) \propto \sum_{c} \underbrace{r_p^{(c)} r_{q'}^{(c)}}_{\text{Amplitude}} \cos\!\left( \underbrace{\Psi_c}_{\text{Content Phase}} + \underbrace{\omega_c \Delta x}_{\text{Spatial Phase}} \right)
87+
$$
88+
89+
📄 **More details in the paper!**
90+
91+
# 🔨 Setup
92+
93+
See the docs folder for detailed setup instructions concerning [installs](docs/INSTALL.md), [datasets](docs/DATASETS.md), [training](docs/TRAINING.md) and [evaluation](docs/EVALUATIONS.md).
94+
95+
Note that: Training NAF takes less than 2 hours consuming less than 8GB of GPU memory on a single NVIDIA A100.
96+
97+
# 📁 Repository Structure
98+
99+
The repository contains the important folders:
100+
```
101+
|-- configs/ # Configuration files
102+
|-- docs/ # Documentation
103+
|-- evaluation/ # Scripts to reproduce results and datasets initialization
104+
|-- notebooks/ # Jupyter notebooks for inference (of any VFM at any scale) and attention map visualizations
105+
|-- src/ # Source code of the project
106+
|-- test/ # Unit tests to compare model efficiency
107+
```
108+
109+
## 🔄 Notebooks
110+
- Inference: [notebooks/inference.ipynb](notebooks/inference.ipynb) runs NAF upsampler on any VFM.
111+
112+
<table align="center">
113+
<tr>
114+
<td align="center">
115+
<img src="asset/inference.png" style="width: 70%;">
116+
<br>
117+
<em>NAF enables zero-shot feature upsampling across any Vision Foundation Model</em>
118+
</td>
119+
</tr>
120+
</table>
121+
<table align="center">
122+
<tr>
123+
<td align="center">
124+
<img src="asset/resolution.png" style="width: 90%;">
125+
<br>
126+
<em>Seamless upsampling from low-resolution to high-resolution features</em>
127+
</td>
128+
</tr>
129+
</table>
130+
131+
- Attention Maps: [notebooks/attention_maps.ipynb](notebooks/attention_maps.ipynb) visualizes NAF neighborhood attention maps.
132+
133+
<table>
134+
<tr>
135+
<td align="center" width="50%">
136+
<img src='./asset/attention.gif' width="90%">
137+
</td>
138+
</tr>
139+
140+
<tr>
141+
<td colspan="2" align="center">
142+
<em>Given a query point and a kernel size, we compute and show its neighborhood attention map.</em>
143+
</td>
144+
</tr>
145+
</table>
146+
147+
## 🔍 Tests
148+
149+
We provide unit tests to evaluate the efficiency of different model configurations, including forward/backward runtime, GPU memory usage, GFLOPs, and number of parameters.
150+
All tests are available in the `test/` directory and we provide the test_results.json file with pre-computed results for reference computed on a 1 A100 40GB GPU.
151+
152+
## 👍 Acknowledgements
153+
154+
Many thanks to these excellent open source projects:
155+
* https://github.com/SHI-Labs/NATTEN
156+
* https://github.com/PaulCouairon/JAFAR
157+
* https://github.com/mhamilton723/FeatUp
158+
* https://github.com/saksham-s/lift/tree/main
159+
* https://github.com/mc-lan/ProxyCLIP
160+
161+
To structure our code we used:
162+
* [Lightning-Hydra Template](https://github.com/ashleve/lightning-hydra-template)
163+
* [Hydra](https://github.com/facebookresearch/hydra)
164+
* [Pytorch-Lightning](https://github.com/Lightning-AI/pytorch-lightning)
165+
166+
Do not hesitate to look and support our previous feature upsampling work:
167+
* https://github.com/PaulCouairon/JAFAR
168+
169+
170+
## ✏️ Bibtex
171+
172+
If this work is helpful for your research, please consider citing the following BibTeX entry and putting a star on this repository. Feel free to open an issue for any questions.
173+
174+
```
175+
@misc{chambon2025naf,
176+
title={NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering.},
177+
author={Loick Chambon and Paul Couairon and Eloi Zablocki and Alexandre Boulch and Nicolas Thome and Matthieu Cord},
178+
year={2025},
179+
}
180+
```

‎asset/architecture.png‎

86 KB
Loading

‎asset/attention.gif‎

5.96 MB
Loading

‎asset/dinov3.png‎

1.77 MB
Loading

‎asset/dog0.jpg‎

5.7 MB
Loading

‎asset/efficiency.png‎

42.9 KB
Loading

‎asset/inference.png‎

1.24 MB
Loading

‎asset/resolution.png‎

1.02 MB
Loading

0 commit comments

Comments
 (0)