research Certified and Robust-by-Design Quality Metrics Provable robustness guarantees and robust architectures for image quality assessment models. 2024–2026 6 papers 7 Adversarial Attacks on Image/Video Quality Metrics New attacks and a large-scale benchmark revealing vulnerabilities of learning-based image and video quality metrics. 2022–2026 6 papers 6 Robust AI-Generated Image Detection (NTIRE 2026 Challenge) Organizing the NTIRE 2026 challenge at CVPR and developing robust, explainable detectors of AI-generated and forged images. 2025–2026 4 papers 13 Robustness of Image Watermarking Benchmarking and attacking invisible watermarks for AI-generated images. 2025–2026 3 papers 14 Physical Adversarial Attacks on Cameras Adversarial patches that work in the real world against video quality metrics and smart-camera systems. 2024–2026 1 paper 3 Attacks on Image Forgery and Deepfake Detectors Anti-forensics attacks that evaluate the reliability of image forgery and deepfake detection. 2026 1 paper 5 Contextual Building Massing Generation (CoMa) Vision-language models that generate site-aware 3D building massings for architectural design. 2026 1 paper 8 Video Quality Metrics Benchmark The largest benchmark of video-compression-related quality metrics, with industry-wide adoption. 2019–2025 5 papers 40 Hacking and Stabilizing VMAF Analysis of how video preprocessing can artificially increase VMAF, the industry-standard video quality metric, and a stable modification of it. 2019–2025 4 papers 10 Defending Quality Metrics Against Adversarial Attacks Benchmarking and developing empirical defenses for image quality metrics, including adversarial purification and adversarial training. 2024–2025 3 papers 13 Robustness of Neural Image Compression Adversarial robustness analysis of learned image codecs, including the JPEG AI standard. 2024–2025 3 papers 7 Video Codec Comparisons and Optimization Methodology and annual comparisons of video codecs and cloud transcoding services. 2019–2021 5 papers 8 Recognition-Aware Video Quality Metrics Development of a new metric predicting object-detection accuracy for compressed videos, in collaboration with Huawei. 2021 Video Conferencing Quality Metrics Methodology and dataset development for video-conferencing quality metrics, in collaboration with Huawei. 2020 Stereoscopic Video Quality and Viewer Discomfort Large-scale study of how 3D video artifacts cause viewer discomfort, and tools for technical quality analysis of stereo films. 2017–2018 3 papers 2