Simultaneous Active Learning of Classifiers & Attributes via Relative Feedback

Abstract

Active learning provides useful tools to reduce annotation costs without compromising classifier performance. However it traditionally views the supervisor simply as a labeling machine. Recently a new interactive learning paradigm was introduced that allows the supervisor to additionally convey useful domain knowledge using attributes. The learner first conveys its belief about an actively chosen image e.g. "I think this is a forest, what do you think?". If the learner is wrong, the supervisor provides an explanation e.g. "No, this is too open to be a forest". With access to a pre-trained set of relative attribute predictors, the learner fetches all unlabeled images more open than the query image, and uses them as negative examples of forests to update its classifier. This rich human-machine communication leads to better classification performance. In this work, we propose three improvements over this set-up. First, we incorporate a weighting scheme that instead of making a hard decision reasons about the likelihood of an image being a negative example. Second, we do away with pre-trained attributes and instead learn the attribute models on the fly, alleviating overhead and restrictions of a pre-determined attribute vocabulary. Finally, we propose an active learning framework that accounts for not just the label-but also the attributes-based feedback while selecting the next query image. We demonstrate significant improvement in classification accuracy on faces and shoes. We also collect and make available the largest relative attributes dataset containing 29 attributes of faces from 60 categories.

Fine-Grained Visual Comparisons with Local Learning

“…In image search, the user could supply semantic feedback to pinpoint his desired content: "the shoes I want to buy are like these but more masculine" [21]. For object recognition, human supervisors could teach the system by relating new objects to previously learned ones, e.g., "a mule has a tail longer than a donkey's" [5,28,30]. In texture recognition, relative attributes could capture the strength of base properties [26].…”

Active Learning for Human Pose Estimation

“…Methods that intelligently design the query space [39,32,30] also share the spirit of reducing annotation effort. Other works have looked into active learning schemes that query for multiple types of annotator feedback [50,4,43]. In this paper, we propose a new computer assisted annotation interface for human pose estimation.…”

Multi-level Adaptive Active Learning for Scene Classification

“…[30] treats the overall object classification problem as a multi-instance learning problem and considers the same type of labels at two levels, instance level (segments) and bag level (images). These works [18,30] nevertheless are still limited to exploiting the same type of standard queries, while another few works [1,21,27,11] have exploited semantic or multiple types of queries. [1,21] introduces a new interactive learning paradigm that allows the supervisor to additionally convey useful domain knowledge using relative attributes.…”

“…These works [18,30] nevertheless are still limited to exploiting the same type of standard queries, while another few works [1,21,27,11] have exploited semantic or multiple types of queries. [1,21] introduces a new interactive learning paradigm that allows the supervisor to additionally convey useful domain knowledge using relative attributes. [27] presents an active learning framework to simultaneously learn appearance and contextual models for scene understanding.…”