Object-Level Backdoor Attacks in RGB-T Semantic Segmentation with Cross-Modality Trigger Optimization
Authors: Xianghao Jiao, Di Wang, Jiawei Liang, Jianjie Huang, Wei Wang, Xiaochun Cao
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments show that the proposed OBA enables precise manipulation of the designated object within the specific class. Extensive experimental results validate our OBA paradigm and CMTG scheme. |
| Researcher Affiliation | Academia | 1School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-Sen University, China 2Peng Cheng Laboratory, Shenzhen, China 3School of Software Technology, Dalian University of Technology, China EMAIL, EMAIL, EMAIL, EMAIL, |
| Pseudocode | Yes | Algorithm 1: Precise Data Poisoning (PDP) |
| Open Source Code | No | The paper does not contain any explicit statement about releasing source code or a link to a code repository. |
| Open Datasets | Yes | We adopt two popular datasets (e.g., MFNet [Ha et al., 2017b] and FMB [Liu et al., 2023a]) to perform the evaluation. |
| Dataset Splits | Yes | MFNet dataset includes 9 classes, in which the background class is labeled 0. Its training, validation and testing sets contain 784, 392 and 393 pairs of images, respectively. FMB dataset describes complex urban street scenes in various severe conditions, e.g., dense fog, heavy rain, and low light. It contains images with 15 categories, and its training and testing sets include 1,220 and 280 pairs, respectively. |
| Hardware Specification | Yes | The training is conducted on a single RTX3090. |
| Software Dependencies | No | The paper mentions EAEFNet as a suitable baseline model and its loss function components (LDice and LSCE), but does not specify any software names with version numbers for implementation. |
| Experiment Setup | Yes | During training, images are resized to 480 640, with a batch size of 8. All other hyperparameters remain consistent with the original paper [Liang et al., 2023]. |