2018/11/21 by Christian Wilms, Wilms, Christian, Simone Frintrop +1
Computer Science · #Advanced Image and Video Retrieval Techniques #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Multimodal Machine Learning Applications
paper · pdf · doi:10.48550/arxiv.1811.08728
openalex publication_date 2018/11/21 · openalex created_date 2022/08/01 · openalex updated_date 2026/07/28
We propose a novel approach for class-agnostic object proposal generation,\nwhich is efficient and especially well-suited to detect small objects.\nEfficiency is achieved by scale-specific objectness attention maps which focus\nthe processing on promising parts of the image and reduce the amount of sampled\nwindows strongly. This leads to a system, which is 33 % faster than the\nstate-of-the-art and clearly outperforming state-of-the-art in terms of average\nrecall. Secondly, we add a module for detecting small objects, which are often\nmissed by recent models. We show that this module improves the average recall\nfor small objects by about 53 %.\n