arXiv Computer Vision research summaries for March 9, 2024.
Today's Research Themes (AI-Generated):
• Novel weakly supervised change detection technique using Knowledge Distillation and Multiscale Sigmoid Inference improves remote sensing applications.
• Self-supervised convolutional neural network demonstrates superior performance in image watermark removal task.
• Adaptive Multi-modal Fusion with Spatially Variant Kernel Refinement enhances diffusion models for realistic image super-resolution.
• Recurrent Aligned Network addresses the domain shift problem in pedestrian trajectory prediction and generalizes to unseen domains.
• SAFDNet, a simple yet effective network for fully sparse 3D object detection, showcases scalability and speed in autonomous driving contexts.
