Joint Segmentation and Recognition of Categorized Objects From Noisy Web Image Collection

Abstract

The segmentation of categorized objects addresses the problem of joint segmentation of a single category of object across a collection of images, where categorized objects are referred to objects in the same category. Most existing methods of segmentation of categorized objects made the assumption that all images in the given image collection contain the target object. In other words, the given image collection is noise free. Therefore, they may not work well when there are some noisy images which are not in the same category, such as those image collections gathered by a text query from modern image search engines. To overcome this limitation, we propose a method for automatic segmentation and recognition of categorized objects from noisy Web image collections. This is achieved by cotraining an automatic object segmentation algorithm that operates directly on a collection of images, and an object category recognition algorithm that identifies which images contain the target object. The object segmentation algorithm is trained on a subset of images from the given image collection which are recognized to contain the target object with high confidence, while training the object category recognition model is guided by the intermediate segmentation results obtained from the object segmentation algorithm. This way, our co-training algorithm automatically identifies the set of true positives in the noisy Web image collection, and simultaneously extracts the target objects from all the identified images. Extensive experiments validated the efficacy of our proposed approach on four datasets: 1) the Weizmann horse dataset, 2) the MSRC object category dataset, 3) the iCoseg dataset, and 4) a new 30-categories dataset including 15,634 Web images with both hand-annotated category labels and ground truth segmentation labels. It is shown that our method compares favorably with the state-of-the-art, and has the ability to deal with noisy image collections.

An Artificial Immune System Classification for Recognition of Categorized Objects from Noisy Web Image Collection

“…The fast HIKSVM proposed in [5] is used to train the object category classifier. The accuracy of the segmentation method [6] is lower than [8], and this may be explained by the fact that [8] has strong ability of encouraging segmentation of images along boundaries of homogeneous color/texture. So incorporate such constraints in the proposed framework.…”

“…There is considerable previous work on Joint segmentation and recognition of categorized objects from noisy web image collection [6]. The algorithm automatically identifies the set of true positives in the noisy Web image collection, and simultaneously extracts the target objects from all the identified images.…”

Video Object Discovery and Co-Segmentation with Extremely Weak Supervision

“…The ith frame of V n with N n i superpixels means that s n ij belongs to the target object co-segmentation methods that conduct the co-segmentation of noisy image collections [25,38], in which several images do not contain the target objects. In our work, we focus on video object discovery and co-segmentation with noisy video collections, where many frames may not contain the target objects.…”

Co-Interest Person Detection from Multiple Wearable Camera Videos

“…Related to this paper is a series of prior research on video co-segmentation, where common objects are segmented from multiple videos. Video co-segmentation can be treated as an extension of the long-studied image cosegmentation [5,11,12,14,17,18,22,25,26,30], where the input is a set of images instead of videos.…”