Segmentation Benchmark
Segmentation benchmarks evaluate the performance of image and video segmentation models, aiming to objectively compare different algorithms and architectures. Current research focuses on improving segmentation accuracy and robustness across diverse datasets and scenarios, often employing transformer-based models, diffusion models, and techniques like masked autoencoders and self-supervised learning. These benchmarks are crucial for advancing the field of computer vision, enabling the development of more accurate and reliable segmentation models with applications in medical image analysis, autonomous driving, and other areas requiring precise object identification and delineation.
Papers
Mitigating Distributional Shift in Semantic Segmentation via Uncertainty Estimation from Unlabelled Data
David S. W. Williams, Daniele De Martini, Matthew Gadd, Paul Newman
Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling
David S. W. Williams, Matthew Gadd, Paul Newman, Daniele De Martini