Cyclic Guidance for Weakly Supervised Joint Detection and Segmentation

Shen, Yunhang; Ji, Rongrong; Wang, Yan; Wu, Yongjian; Cao, Liujuan

doi:10.1109/cvpr.2019.00079

Cited by 111 publications

(76 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Fang et al [64] developed an object detection model that solves the non-convexity problem with a series of smoothed loss functions, and proved the good overall performance of the model, as well as its advantage in positioning. Based on semantic segmentation, Shen et al [65] proposed an object detection model with a multi-tasking learning mechanism, and observed that the model attained competitive outcomes against other alternatives. Yang et al [66] developed an image object positioning model, which, unlike the earlier approaches, is immune to local minimum trap.…”

Section: Literature Reviewmentioning

confidence: 99%

A Region-Based Efficient Network for Accurate Object Detection

Guan¹,

Aamir²,

Hu³

et al. 2021

View full text Add to dashboard Cite

Object detection in images is an important task in image processing and computer vision. Many approaches are available for object detection. For example, there are numerous algorithms for object positioning and classification in images. However, the current methods perform poorly and lack experimental verification. Thus, it is a fascinating and challenging issue to position and classify image objects. Drawing on the recent advances in image object detection, this paper develops a region-baed efficient network for accurate object detection in images. To improve the overall detection performance, image object detection was treated as a twofold problem, involving object proposal generation and object classification. First, a framework was designed to generate high-quality, class-independent, accurate proposals. Then, these proposals, together with their input images, were imported to our network to learn convolutional features. To boost detection efficiency, the number of proposals was reduced by a network refinement module, leaving only a few eligible candidate proposals. After that, the refined candidate proposals were loaded into the detection module to classify the objects. The proposed model was tested on the test set of the famous PASCAL Visual Object Classes Challenge 2007 (VOC2007). The results clearly demonstrate that our model achieved robust overall detection efficiency over existing approaches using fewer or more proposals, in terms of recall, mean average best overlap (MABO), and mean average precision (mAP).

show abstract

Section: Literature Reviewmentioning

confidence: 99%

A Region-Based Efficient Network for Accurate Object Detection

Guan¹,

Aamir²,

Hu³

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Building upon WSDDN, Zhang et al [58] propose a zigzag strategy to discover reliable object instances and train them by curriculum learning [2]. Shen et al [38] jointly learn weakly supervised detection and segmentation, using failures of one task to complement the other. Tang et al [48] extend WSDDN to multiple instance detection network including online instance classifier refinement (OICR) and introduce a weaklysupervised region proposal network as a plugin [49].…”

Section: Related Workmentioning

confidence: 99%

Training object detectors from few weakly-labeled and many unlabeled images

Yang

Shi

et al. 2021

Pattern Recognition

View full text Add to dashboard Cite

Weakly-supervised object detection attempts to limit the amount of supervision by dispensing the need for bounding boxes, but still assumes image-level labels on the entire training set are available. In this work, we study the problem of training an object detector from one or few clean images with image-level labels and a larger set of completely unlabeled images. This is an extreme case of semisupervised learning where the labeled data are not enough to bootstrap the learning of a classifier or detector.Our solution is to use a standard weakly-supervised pipeline to train a student model from image-level pseudolabels generated on the unlabeled set by a teacher model, bootstrapped by region-level similarities to clean labeled images. By using the recent pipeline of PCL [47] and more unlabeled images, we achieve performance competitive or superior to many state of the art weakly-supervised detection solutions.

show abstract

“…In addition, for most non-rigid objects ("cat", "dog", "horse", "person', etc. ), as can be seen from Table 1, by applying in- (Tang et al 2017) 60.6 PCL (Tang et al 2018) 62.7 C-WSL* (Gao et al 2018) 63.5 MELM (Wan et al 2018) 61.4 WS-JDS (Shen et al 2019) 64.5 C-MIL (Wan et al 2019) 65.0 OICR+W-RPN (Singh and Lee 2019) 66.5 SDCN (Li et al 2019) 66.8 OIM+IR 67.2 C-WSL*+FRCNN (Gao et al 2018) 66.1 WS-JDS+FRCNN (Shen et al 2019) 68.6 SDCN+FRCNN (Li et al 2019)…”

Section: Comparison With State-of-the-artsmentioning

confidence: 99%

Object Instance Mining for Weakly Supervised Object Detection

Lin¹,

Wang

Xu³

et al. 2020

AAAI

View full text Add to dashboard Cite

Weakly supervised object detection (WSOD) using only image-level annotations has attracted growing attention over the past few years. Existing approaches using multiple instance learning easily fall into local optima, because such mechanism tends to learn from the most discriminative object in an image for each category. Therefore, these methods suffer from missing object instances which degrade the performance of WSOD. To address this problem, this paper introduces an end-to-end object instance mining (OIM) framework for weakly supervised object detection. OIM attempts to detect all possible object instances existing in each image by introducing information propagation on the spatial and appearance graphs, without any additional annotations. During the iterative learning process, the less discriminative object instances from the same class can be gradually detected and utilized for training. In addition, we design an object instance reweighted loss to learn larger portion of each object instance to further improve the performance. The experimental results on two publicly available databases, VOC 2007 and 2012, demonstrate the efficacy of proposed approach.

show abstract

Cyclic Guidance for Weakly Supervised Joint Detection and Segmentation

Cited by 111 publications

References 55 publications

A Region-Based Efficient Network for Accurate Object Detection

A Region-Based Efficient Network for Accurate Object Detection

Training object detectors from few weakly-labeled and many unlabeled images

Object Instance Mining for Weakly Supervised Object Detection

Contact Info

Product

Resources

About