Fully supervised object detection requires training images in which all instances are annotated. This is actually impractical due to the high labor and time costs unavoidable missing annotations. As a result, incomplete annotation each image could provide misleading supervision harm training. Recent works on sparsely annotated alleviate this problem by generating pseudo labels for Such mechanis...

Abstract Aiming at the problems of low detection accuracy and blurred object edges in current salient based on background algorithms, a new algorithm boundary prior to estimate is proposed. Firstly, super-pixel image segmentation (SLIC) used segment into super-pixels. According theory that foreground are relative, estimated by using four-edges image; At same time, an optimal fusion proposed fus...

Moving object detection from aerial images remains an unsolved problem in computer vision research domain. Detection results are not precise due to blurry images, thin edges and noise. Various methods were previously proposed for moving which could provide robust many challenges, i.e., noise, motion detection, lack of appropriate features, effective classification approach, complex background v...

Object detection, as of one the most fundamental and challenging problems in computer vision, has received great attention recent years. Over past two decades, we have seen a rapid technological evolution object detection its profound impact on entire vision field. If consider today’s technique revolution driven by deep learning, then, back 1990s, would see ingenious thinking long-term perspect...

RoIPool/RoIAlign is an indispensable process for the typical two-stage object detection algorithm, it used to rescale proposal cropped from feature pyramid generate a fixed size map. However, these maps of local receptive fields will heavily lose global context information. To tackle this problem, we propose novel end-to-end trainable framework, called aware (GCA) RCNN, aiming at assisting neur...

Few-shot object detection (FSOD) eliminates the dependence on tremendous instances with manual annotations in conventional detection. We deem that scarcity of positive samples is main reason restricts performance FSOD detectors. In this paper, a novel model via sample processing, namely, FSSP, proposed to detect objects accurately only few annotated samples, which based structural design Siames...

Abstract It is an important task to estimate a 3D bounding box from monocular images for autonomous driving. However, the pictures do not have distance information, so it difficult acquire accurate results. For sake of solving trouble low accuracy image in target detection because lacking improved three-dimensional algorithm based on GUPNet and neural network was proposed promote precision dete...

Intelligent detection of marine organism plays an important part in the economy, and it is significant to detect organisms quickly accurately a complex environment for intelligence equipment. The existing object models do not work well underwater. This paper improves structure EfficientDet detector proposes EfficientDet-Revised (EDR), which new model. Specifically, MBConvBlock reconstructed by ...

Object Detection is one of the most popular applications in branch computer vision. While accuracy has always been focus, focus gradually also shifted to lightweight models. In this paper we propose a light weight system where object classification not required but only detection using clustering methods.

For autonomy in the maritime domain, object detection is a very important task, as one needs to perceive surroundings take appropriate action decisions. A large issue and classification shortage of thorough datasets. In this work, our aim reduce problem by introducing pipeline for generation simulated data that matches target thereby achieving more reliable robust performance detector. This eas...