The existing approaches for salient motion segmentation are unable to explicitly learn geometric cues and often give false detections on prominent static objects. We exploit multiview geometric constraints to avoid such shortcomings. To handle the nonrigid background like a sea, we also propose a robust fusion mechanism between motion and appearance-based features. We find dense trajectories, covering every pixel in the video, and propose trajectory-based epipolar distances to distinguish between background and foreground regions. Trajectory epipolar distances are data-independent and can be readily computed given a few features' correspondences between the images. We show that by combining epipolar distances with optical flow, a powerful motion network can be learned. Enabling the network to leverage both of these features, we propose a simple mechanism, we call input-dropout. Comparing the motion-only networks, we outperform the previous state of the art on DAVIS-2016 dataset by 5.2% in the mean IoU score. By robustly fusing our motion network with an appearance network using the input-dropout mechanism, we also outperform the previous methods on DAVIS-2016, 2017 and Segtrackv2 dataset.
Exploiting saliency for object segmentation from image level labels
Max Planck Institute for Informatics(马克斯·普朗克计算机科学研究所);Amsterdam Machine Learning Lab(阿姆斯特丹机器学习实验室);Saarland Informatics Campus(萨尔兰信息学园区);University of Amsterdam(阿姆斯特丹大学)
City University of Hong Kong(香港城市大学);Shenyang Institute of Automation, Chinese Academy of Sciences(中国科学院沈阳自动化研究所);University of Chinese Academy of Sciences(中国科学院大学)
Siamese Network for RGB-D Salient Object Detection and Beyond
Sichuan University(四川大学);National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University(四川大学视觉合成图形图像技术国家级重点实验室);Nankai University(南开大学);Wuhan University(武汉大学);Inception Institute of Artificial Intelligence (IIAI)(起源人工智能研究院);University of Electronic Science and Technology of China(电子科技大学)
Alignment-Free RGBT Salient Object Detection: Semantics-guided Asymmetric Correlation Network and A Unified Benchmark
Anhui University(安徽大学);School of Computer Science and Technology(计算机科学与技术学院);Anhui Provincial Key Laboratory of Multimodal Cognitive Computation(安徽省多模态认知计算重点实验室);Information Materials and Intelligent Sensing Laboratory of Anhui Province(安徽省信息材料与智能感知实验室)