8 papers
ASBench: Image Anomalies Synthesis Benchmark for Anomaly Detection
Qunyi Zhang, Songan Zhang, Jiaqi Liu +5
Anomaly detection plays a pivotal role in manufacturing quality control, yet its application is constrained by limited abnormal samples and high manual annotation costs. While anom…
FAST: Foreground-aware Diffusion with Accelerated Sampling Trajectory for Segmentation-oriented Anomaly Synthesis
Xichen Xu, Yanshu Wang, Jinbao Wang +4
Industrial anomaly segmentation relies heavily on pixel-level annotations, yet real-world anomalies are often scarce, diverse, and costly to label. Segmentation-oriented industrial…
A Survey on Industrial Anomalies Synthesis
Yanshu Wang, Xichen Xu, Jiaqi Liu +4
This paper comprehensively reviews anomaly synthesis methodologies. Existing surveys focus on limited techniques, missing an overall field view and understanding method interconnec…
STAGE: Segmentation-oriented Industrial Anomaly Synthesis via Graded Diffusion with Explicit Mask Alignment
Xichen Xu, Yanshu Wang, Jinbao Wang +5
Segmentation-oriented Industrial Anomaly Synthesis (SIAS) plays a pivotal role in enhancing the performance of downstream anomaly segmentation, as it provides an effective means of…
Trade-offs in Image Generation: How Do Different Dimensions Interact?
Sicheng Zhang, Binzhu Xie, Zhonghao Yan +7
Model performance in text-to-image (T2I) and image-to-image (I2I) generation often depends on multiple aspects, including quality, alignment, diversity, and robustness. However, mo…
Learning from Loss Landscape: Generalizable Mixed-Precision Quantization via Adaptive Sharpness-Aware Gradient Aligning
Lianbo Ma, Jianlun Ma, Yuee Zhou +3
Mixed Precision Quantization (MPQ) has become an essential technique for optimizing neural network by determining the optimal bitwidth per layer. Existing MPQ methods, however, fac…