Diffract and Conquer: Hyperspectral Imaging from Any RGB Camera via Optical Encoding and Learning

Pronin, Alexey; Vladimirov, Daniil; Korepanov, Andrei; Muzyka, Artem; Makarov, Andrey; Podtikhova, Sofya; Ershov, Egor; Rastorguev, Andrey; Skidanov, Roman; Nikonorov, Artem

doi:10.24963/ijcai.2026/170

Diffract and Conquer: Hyperspectral Imaging from Any RGB Camera via Optical Encoding and Learning

Alexey Pronin, Daniil Vladimirov, Andrei Korepanov, Artem Muzyka, Andrey Makarov, Sofya Podtikhova, Egor Ershov, Andrey Rastorguev, Roman Skidanov, Artem Nikonorov

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 1523-1531.

https://doi.org/10.24963/ijcai.2026/170

Keywords

Computer Vision: Low-level Vision
Computer Vision: Applications and Systems
Computer Vision: Computational photography

Abstract

Hyperspectral imaging (HSI) provides detailed spectral information but is often impractical due to high cost, size, and hardware complexity. While learning-based methods attempt to recover hyperspectral images from RGB sensors, they are constrained by limited spectral measurements and noise. We propose a unified optical–computational approach that converts any standard RGB camera into a snapshot hyperspectral imaging system. Our method introduces a diffractive optical adapter that replaces the conventional lens with a diffractive lens array optimized for spectral encoding. To reconstruct hyperspectral images from the resulting measurements, we design a neural network specialized for diffractively encoded RGB data, capable of compensating for optical distortions and recovering high-quality spectra. The proposed system achieves up to 10 dB improvement in PSNR over RGB-based hyperspectral reconstruction and enhances the performance of existing state-of-theart models by up to 6 dB when used with the proposed adapter. Our results demonstrate that diffractive optical encoding combined with learned reconstruction enables practical and scalable hyperspectral imaging using commodity RGB cameras.

Alexey Pronin, Daniil Vladimirov, Andrei Korepanov, Artem Muzyka, Andrey Makarov, Sofya Podtikhova, Egor Ershov, Andrey Rastorguev, Roman Skidanov, and Artem Nikonorov. 2026. Diffract and Conquer: Hyperspectral Imaging from Any RGB Camera via Optical Encoding and Learning. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 1523–1531. https://doi.org/10.24963/ijcai.2026/170

@inproceedings{ijcai2026p170,
  title     = {Diffract and Conquer: Hyperspectral Imaging from Any RGB Camera via Optical Encoding and Learning},
  author    = {Pronin, Alexey and Vladimirov, Daniil and Korepanov, Andrei and Muzyka, Artem and Makarov, Andrey and Podtikhova, Sofya and Ershov, Egor and Rastorguev, Andrey and Skidanov, Roman and Nikonorov, Artem},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {1523--1531},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/170},
  url       = {https://doi.org/10.24963/ijcai.2026/170},
}

TY  - CPAPER
TI  - Diffract and Conquer: Hyperspectral Imaging from Any RGB Camera via Optical Encoding and Learning
AU  - Pronin, Alexey
AU  - Vladimirov, Daniil
AU  - Korepanov, Andrei
AU  - Muzyka, Artem
AU  - Makarov, Andrey
AU  - Podtikhova, Sofya
AU  - Ershov, Egor
AU  - Rastorguev, Andrey
AU  - Skidanov, Roman
AU  - Nikonorov, Artem
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 1523
EP  - 1531
DO  - 10.24963/ijcai.2026/170
UR  - https://doi.org/10.24963/ijcai.2026/170
ER  -