Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Spectral Compressive Imaging via Chromaticity-Intensity Decomposition
Authors: Xiaodong Wang, Zijun He, Ping Wang, Lishun Wang, Yanan Hu, Xin Yuan
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on both synthetic and real-world CASSI datasets demonstrate that our method achieves superior performance in both spectral and chromaticity fidelity. Code is released at: https://github.com/xiaodongwo/CIDNet. |
| Researcher Affiliation | Academia | Xiaodong Wang 1,2, Zijun He 1,2, Ping Wang 2, Lishun Wang 3, Yanan Hu 1,2, Xin Yuan 2, 1 Zhejiang University, 2 School of Engineering, Westlake University, 3 Chengdu Institute of Biology, Chinese Academy of Sciences |
| Pseudocode | No | The paper describes the iterative optimization process using mathematical equations (Eq. 11-18) and text, but it does not present a structured pseudocode or algorithm block. |
| Open Source Code | Yes | Code is released at: https://github.com/xiaodongwo/CIDNet. |
| Open Datasets | Yes | We adopt two widely used hyperspectral datasets: CAVE [31] and KAIST [6]. |
| Dataset Splits | Yes | Following prior works [20, 21, 3], we use all CAVE images for training and select 10 scenes from KAIST for evaluation. |
| Hardware Specification | Yes | All experiments are conducted in Nvidia A40 GPU. |
| Software Dependencies | No | Our model is implemented in Py Torch and trained using the Adam optimizer [13] for 300 epochs. The text mentions PyTorch but does not specify a version number. |
| Experiment Setup | Yes | Our model is implemented in Py Torch and trained using the Adam optimizer [13] for 300 epochs. The initial learning rate is set to 4 10 4 and updated using a cosine annealing schedule. We employ the ℓ2 loss between the reconstructed chromaticity and ground-truth chromaticity as the objective function. For training, we randomly extract 3D hyperspectral patches from each scene. For simulated data, the patch size is 256 256 28, for real-world data, we use patches of size 350 260 26. |