We consider the task of 3-d depth estimation from a single still image. We take a supervised learning approach to this problem, in which we begin by collecting a training set of monocular images (of unstructured indoor and outdoor environments which include forests, sidewalks, trees, buildings, etc.) and their corresponding ground-truth depthmaps. Then, we apply supervised learning to predict the value of the depthmap as a function of the image. Depth estimation is a challenging problem, since local features alone are insufficient to estimate depth at a point, and one needs to consider the global context of the image. Our model uses a hierarchical, multiscale Markov Random Field (MRF) that incorporates multiscale local-and global-image features, and models the depths and the relation between depths at different points in the image. We show that, even on unstructured scenes, our algorithm is frequently able to recover fairly accurate depthmaps. We further propose a model that incorporates both monocular cues and stereo (triangulation) cues, to obtain significantly more accurate depth estimates than is possible using either monocular or stereo cues alone.

No abstract

Product search engine faces unique challenges that differ from web page search. The goal of a product search engine is to rank relevant items that the user may be interested in purchasing. Clicks provide a strong signal of a user's interest in an item. Traditional click prediction models include many features such as document text, price, and user information. In this paper, we propose adding information extracted from the thumbnail image of the item as additional features for click prediction. Specifically, we use two types of image features -photographic features and object features. Our experiments reveal that both types of features can be highly useful in click prediction. We measure our performance in both prediction accuracy and NDCG. Overall, our experiments show that augmenting with image features to a standard model in click prediction provides significant improvement in precision and recall and boosts NDCG.

No abstract

No abstract