3 papers
cs.CV2025
Flashback: Memory-Driven Zero-shot, Real-time Video Anomaly Detection
Hyogun Lee, Haksub Kim, Ig-Jae Kim +1
Video Anomaly Detection (VAD) automatically identifies anomalous events from video, mitigating the need for human operators in large-scale surveillance deployments. However, two fu…
cs.CV2025
Navigating Label Ambiguity for Facial Expression Recognition in the Wild
JunGyu Lee, Yeji Choi, Haksub Kim +2
Facial expression recognition (FER) remains a challenging task due to label ambiguity caused by the subjective nature of facial expressions and noisy samples. Additionally, class i…
cs.CV2024
IG-FIQA: Improving Face Image Quality Assessment through Intra-class Variance Guidance robust to Inaccurate Pseudo-Labels
Minsoo Kim, Gi Pyo Nam, Haksub Kim +2
In the realm of face image quality assesment (FIQA), method based on sample relative classification have shown impressive performance. However, the quality scores used as pseudo-la…