HSOD-BIT-V2: A Challenging Benchmark for Hyperspectral Salient Object Detection

Authors: Yuhao Qiu, Shuyan Bai, Tingfa Xu, Peifu Liu, Haolin Qin, Jianan Li

AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments have been conducted to evaluate the performance of Hyper-HRNet on HSOD-BIT-V2, HSOD-BIT and HS-SOD datasets. Our model surpasses mainstream models, especially in challenging backgrounds. Experimental analysis demonstrates that Hyper-HRNet outperforms existing models, especially in challenging scenarios.
Researcher Affiliation Academia Yuhao Qiu, Shuyan Bai, Tingfa Xu , Peifu Liu, Haolin Qin, Jianan Li Beijing Institute of Technology
Pseudocode No The paper describes the network architecture and methods in detail, but does not include any explicitly labeled 'Pseudocode' or 'Algorithm' blocks.
Open Source Code No The paper does not contain an unambiguous statement or a direct link to a code repository indicating that the source code for the methodology is openly available. It mentions "Further experiments and details are available in the supplementary material" but does not explicitly state code release.
Open Datasets Yes In this context, we introduce HSOD-BIT-V2, the largest and most challenging HSOD benchmark dataset to date. This dataset includes eight natural scene backgrounds and, for the first time in HSOD, introduces snowfields and fallen leaves, greatly enhancing diversity. The dataset emphasizes spectral advantages through five challenging attributes, focusing on small objects and foreground-background similarities. HSOD-BIT dataset (2024), while larger and including some challenges, still lacks sufficient challenging data to fully showcase spectral advantages. The first tailored HSOD dataset HS-SOD is small and limited to common scenes (2018).
Dataset Splits Yes From the 500 processed data cubes, 406 images were used for training, and 94 for testing.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments (e.g., GPU models, CPU types, or cloud instance specifications).
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., programming languages, libraries, or frameworks with their respective versions).
Experiment Setup No The paper states, "Further experiments and details are available in the supplementary material." and for HS-SOD dataset, "using consistent training configurations from previous works (2023; 2024)" but does not explicitly provide concrete hyperparameter values or detailed training configurations in the main text.