Webly Supervised Semantic Segmentation

Jin, Bin; Segovia, Maria V. Ortiz; Süsstrunk, Sabine

doi:10.1109/cvpr.2017.185

Cited by 66 publications

(46 citation statements)

References 36 publications

(119 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Since the object-level annotations are more accurate and comprehensive than image-level annotations, these methods usually have better performance. For example, the methods trained with accurate bounding boxes have performance more than 60% IOU, while most of the methods with image-level [45] 37.8 37.0 EM-Adapt (ICCV 2015) [17] 38.2 39.6 DCSM (ECCV 2016) [48] 44.1 45.1 BFBP (ECCV 2016) [44] 46.6 48.0 STC (PAMI 2017) [49] 49.8 51.2 SEC (ECCV 2016) [22] 50.7 51.7 AF-SS (ECCV 2016) [19] 52.6 52.7 WebSeg (CVPR 2017) [50] 53.4 55.3 AE-PSL (CVPR 2017) [14] 55.0 55.7 WebCoSeg (BMVC 2017) [51] 56.4 56.9 DSNA (Arxiv 2018) [24] 58.2 60.1 MDC (CVPR 2018) [23] 60.4 60.8 MCOF (CVPR 2018) [52] 60.3 61.2 AffinityNet (CVPR 2018) [53] 61.7 63.7 Boostrap(CVPR 2018) [54] 63.0 63.9 InstancesSalient(ECCV 2018) [55] 64.5 65.6 Ours 61.9 62.8 labels have the performance less than 60%. Some image-level based methods implicitly use pixel-level supervision in their models such as [19], [21], therefore their models can achieve relatively higher performance than those only using imagelevel labels.…”

Section: Semantic Segmentation Resultsmentioning

confidence: 99%

Coarse-to-Fine Semantic Segmentation From Image-Level Labels

Jing

Chen

Tian

2020

IEEE Trans. on Image Process.

105

View full text Add to dashboard Cite

Deep neural network-based semantic segmentation generally requires large-scale cost extensive annotations for training to obtain better performance. To avoid pixel-wise segmentation annotations which are needed for most methods, recently some researchers attempted to use object-level labels (e.g. bounding boxes) or image-level labels (e.g. image categories). In this paper, we propose a novel recursive coarse-to-fine semantic segmentation framework based on only image-level category labels. For each image, an initial coarse mask is first generated by a convolutional neural network-based unsupervised foreground segmentation model and then is enhanced by a graph model. The enhanced coarse mask is fed to a fully convolutional neural network to be recursively refined. Unlike existing image-level label-based semantic segmentation methods which require to label all categories for images contain multiple types of objects, our framework only needs one label for each image and can handle images contains multi-category objects. With only trained on ImageNet, our framework achieves comparable performance on PASCAL VOC dataset as other image-level label-based state-ofthe-arts of semantic segmentation. Furthermore, our framework can be easily extended to foreground object segmentation task and achieves comparable performance with the state-of-the-art supervised methods on the Internet Object dataset.

show abstract

Section: Semantic Segmentation Resultsmentioning

confidence: 99%

Coarse-to-Fine Semantic Segmentation From Image-Level Labels

Jing

Chen

Tian

2020

IEEE Trans. on Image Process.

105

View full text Add to dashboard Cite

show abstract

“…Weakly Supervised Signals for Segmentation: Numerous alternatives to expensive pixel-level segmentation have been proposed and used in the literature. Image-level labels [27], noisy web labels [1,16] and scribble-level labels [20] are some of the supervisory signal that have been used to guide segmentation methods. Closer to our approach, [3] employs point-level supervision in the form of a single click to train a CNN for semantic segmentation and [26] uses central points of an imaginary bounding box to weakly supervise object detection.…”

Section: Related Workmentioning

confidence: 99%

Deep Extreme Cut: From Extreme Points to Object Segmentation

Maninis

Caelles

Pont-Tuset

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

413

426

View full text Add to dashboard Cite

Figure 1. Example results of DEXTR: The user provides the extreme clicks for an object, and the CNN produces the segmented masks. AbstractThis paper explores the use of extreme points in an object (left-most, right-most, top, bottom pixels) as input to obtain precise object segmentation for images and videos. We do so by adding an extra channel to the image in the input of a convolutional neural network (CNN), which contains a Gaussian centered in each of the extreme points. The CNN learns to transform this information into a segmentation of an object that matches those extreme points.We demonstrate the usefulness of this approach for guided segmentation (grabcut-style), interactive segmentation, video object segmentation, and dense segmentation annotation. We show that we obtain the most precise results to date, also with less user input, in an extensive and varied selection of benchmarks and datasets. All our models and code are publicly available on

show abstract

“…To improve the localization performance, some approaches [58][59][60][61] have proposed to exploit the notion of objectness, either by incorporating it in the loss function [58,59], or by employing pre-trained network as external objectness module [60,61]. Another promising way to improve the segmentation performance is to utilize additional weakly supervised images, such as web images, to train CNNs, such as [62,63].…”

Section: Weakly Supervised Semantic Segmentationmentioning

confidence: 99%

A review of semantic segmentation using deep neural networks

Guo

Liu

Georgiou

et al. 2017

Int J Multimed Info Retr

618

263

View full text Add to dashboard Cite

During the long history of computer vision, one of the grand challenges has been semantic segmentation which is the ability to segment an unknown image into different parts and objects (e.g., beach, ocean, sun, dog, swimmer). Furthermore, segmentation is even deeper than object recognition because recognition is not necessary for segmentation. Specifically, humans can perform image segmentation without even knowing what the objects are (for example, in satellite imagery or medical X-ray scans, there may be several objects which are unknown, but they can still be segmented within the image typically for further investigation). Performing segmentation without knowing the exact identity of all objects in the scene is an important part of our visual understanding process which can give us a powerful model to understand the world and also be used to improve or augment existing computer vision techniques. Herein this work, we review the field of semantic segmentation as pertaining to deep convolutional neural networks. We provide comprehensive coverage of the top approaches and summarize the strengths, weaknesses and major challenges.

show abstract

Webly Supervised Semantic Segmentation

Cited by 66 publications

References 36 publications

Coarse-to-Fine Semantic Segmentation From Image-Level Labels

Coarse-to-Fine Semantic Segmentation From Image-Level Labels

Deep Extreme Cut: From Extreme Points to Object Segmentation

A review of semantic segmentation using deep neural networks

Contact Info

Product

Resources

About