Uncertainty-aware Continuous Implicit Neural Representations for Remote Sensing Object Counting

Siyuan Xu, Yucheng Wang, Mingzhou Fan, Byung-Jun Yoon, Xiaoning Qian
Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, PMLR 238:4105-4113, 2024.

Abstract

Many existing object counting methods rely on density map estimation (DME) of the discrete grid representation by decoding extracted image semantic features from designed convolutional neural networks (CNNs). Relying on discrete density maps not only leads to information loss dependent on the original image resolution, but also has a scalability issue when analyzing high-resolution images with cubically increasing memory complexity. Furthermore, none of the existing methods can offer reliable uncertainty quantification (UQ) for the derived count estimates. To overcome these limitations, we design UNcertainty-aware, hypernetwork-based Implicit neural representations for Counting (UNIC) to assign probabilities and the corresponding counting confidence over continuous spatial coordinates. We derive a sampling-based Bayesian counting loss function and develop the corresponding model training algorithm. UNIC outperforms existing methods on the Remote Sensing Object Counting (RSOC) dataset with reliable UQ and improved interpretability of the derived count estimates. Our code is available at https://github.com/SiyuanXu-tamu/UNIC.

Cite this Paper

BibTeX
@InProceedings{pmlr-v238-xu24b, title = {Uncertainty-aware Continuous Implicit Neural Representations for Remote Sensing Object Counting}, author = {Xu, Siyuan and Wang, Yucheng and Fan, Mingzhou and Yoon, Byung-Jun and Qian, Xiaoning}, booktitle = {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics}, pages = {4105--4113}, year = {2024}, editor = {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen}, volume = {238}, series = {Proceedings of Machine Learning Research}, month = {02--04 May}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v238/xu24b/xu24b.pdf}, url = {https://proceedings.mlr.press/v238/xu24b.html}, abstract = {Many existing object counting methods rely on density map estimation (DME) of the discrete grid representation by decoding extracted image semantic features from designed convolutional neural networks (CNNs). Relying on discrete density maps not only leads to information loss dependent on the original image resolution, but also has a scalability issue when analyzing high-resolution images with cubically increasing memory complexity. Furthermore, none of the existing methods can offer reliable uncertainty quantification (UQ) for the derived count estimates. To overcome these limitations, we design UNcertainty-aware, hypernetwork-based Implicit neural representations for Counting (UNIC) to assign probabilities and the corresponding counting confidence over continuous spatial coordinates. We derive a sampling-based Bayesian counting loss function and develop the corresponding model training algorithm. UNIC outperforms existing methods on the Remote Sensing Object Counting (RSOC) dataset with reliable UQ and improved interpretability of the derived count estimates. Our code is available at https://github.com/SiyuanXu-tamu/UNIC.} }
Endnote
%0 Conference Paper %T Uncertainty-aware Continuous Implicit Neural Representations for Remote Sensing Object Counting %A Siyuan Xu %A Yucheng Wang %A Mingzhou Fan %A Byung-Jun Yoon %A Xiaoning Qian %B Proceedings of The 27th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2024 %E Sanjoy Dasgupta %E Stephan Mandt %E Yingzhen Li %F pmlr-v238-xu24b %I PMLR %P 4105--4113 %U https://proceedings.mlr.press/v238/xu24b.html %V 238 %X Many existing object counting methods rely on density map estimation (DME) of the discrete grid representation by decoding extracted image semantic features from designed convolutional neural networks (CNNs). Relying on discrete density maps not only leads to information loss dependent on the original image resolution, but also has a scalability issue when analyzing high-resolution images with cubically increasing memory complexity. Furthermore, none of the existing methods can offer reliable uncertainty quantification (UQ) for the derived count estimates. To overcome these limitations, we design UNcertainty-aware, hypernetwork-based Implicit neural representations for Counting (UNIC) to assign probabilities and the corresponding counting confidence over continuous spatial coordinates. We derive a sampling-based Bayesian counting loss function and develop the corresponding model training algorithm. UNIC outperforms existing methods on the Remote Sensing Object Counting (RSOC) dataset with reliable UQ and improved interpretability of the derived count estimates. Our code is available at https://github.com/SiyuanXu-tamu/UNIC.
APA
Xu, S., Wang, Y., Fan, M., Yoon, B. & Qian, X.. (2024). Uncertainty-aware Continuous Implicit Neural Representations for Remote Sensing Object Counting. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 238:4105-4113 Available from https://proceedings.mlr.press/v238/xu24b.html.

Related Material