This paper proposes a conceptually simple but surprisingly powerful method which combines the effectiveness of a discriminative object detector with the explicit correspondence offered by a nearest-neighbor approach. The method is based on training a separate linear SVM classifier for every exemplar in the training set. Each of these Exemplar-SVMs is thus defined by a single positive instance and millions of negatives. While each detector is quite specific to its exemplar, we empirically observe that an ensemble of such Exemplar-SVMs offers surprisingly good generalization. Our performance on the PASCAL VOC detection task is on par with the much more complex latent part-based model of Felzenszwalb et al., at only a modest computational cost increase. But the central benefit of our approach is that it creates an explicit association between each detection and a single training exemplar. Because most detections show good alignment to their associated exemplar, it is possible to transfer any available exemplar meta-data (segmentation, geometric structure, 3D model, etc.) directly onto the detections, which can then be used as part of overall scene understanding. 1.

Abstract. The Deformable Parts Model (DPM) has recently emerged as a very useful and popular tool for tackling the intra-category diversity problem in object detection. In this paper, we summarize the key insights from our empirical analysis of the important elements constituting this detector. More specifically, we study the relationship between the role of deformable parts and the mixture model components within this detector, and understand their relative importance. First, we find that by increasing the number of components, and switching the initialization step from their aspect-ratio, left-right flipping heuristics to appearancebased clustering, considerable improvement in performance is obtained. But more intriguingly, we observed that with these new components, the part deformations can now be turned off, yet obtaining results that are almost on par with the original DPM detector.

Abstract not found

We present a unified model for face detection, pose estimation, and landmark estimation in real-world, cluttered images. Our model is based on a mixtures of trees with a shared pool of parts; we model every facial landmark as a part and use global mixtures to capture topological changes due to viewpoint. We show that tree-structured models are surprisingly effective at capturing global elastic deformation, while being easy to optimize unlike dense graph structures. We present extensive results on standard face benchmarks, as well as a new “in the wild ” annotated dataset, that suggests our system advances the state-of-theart, sometimes considerably, for all three tasks. Though our model is modestly trained with hundreds of faces, it compares favorably to commercial systems trained with billions of examples (such as Google Picasa and face.com). 1.

While there has been a lot of recent work on object recognition and image understanding, the focus has been on carefully establishing mathematical models for images, scenes, and objects. In this paper, we propose a novel, nonparametric approach for object recognition and scene parsing using a new technology we name label transfer. For an input image, our system first retrieves its nearest neighbors from a large database containing fully annotated images. Then, the system establishes dense correspondences between the input image and each of the nearest neighbors using the dense SIFT flow algorithm [28], which aligns two images based on local image structures. Finally, based on the dense scene correspondences obtained from SIFT flow, our system warps the existing annotations and integrates multiple cues in a Markov random field framework to segment and recognize the query image. Promising experimental results have been achieved by our nonparametric scene parsing system on challenging databases. Compared to existing object recognition approaches that require training classifiers or appearance models for each object category, our system is easy to implement, has few parameters, and embeds contextual information naturally in the retrieval/alignment procedure.

the date of receipt and acceptance should be inserted later Abstract Given a single outdoor image, we present a method for estimating the likely illumination conditions of the scene. In particular, we compute the probability distribution over the sun position and visibility. The method relies on a combination of weak cues that can be extracted from different portions of the image: the sky, the vertical surfaces, the ground, and the convex objects in the image. While no single cue can reliably estimate illumination by itself, each one can reinforce the others to yield a more robust estimate. This is combined with a data-driven prior computed over a dataset of 6 million photos. We present quantitative results on a webcam dataset with annotated sun positions, as well as quantitative and qualitative results on consumer-grade photographs downloaded from Internet. Based on the estimated illumination, we show how to realistically insert synthetic 3-D objects into the scene, and how to transfer appearance across images while keeping the illumination consistent. Keywords illumination estimation · data-driven methods · shadow detection · scene understanding · image synthesis 1

Abstract Many state-of-the-art approaches for object recognition reduce the problem to a 0-1 classification task. This allows one to leverage sophisticated machine learning techniques for training classifiers from labeled examples. However, these models are typically trained independently for each class using positive and negative examples cropped from images. At test-time, various post-processing heuristics such as non-maxima suppression (NMS) are required to reconcile multiple detections within and between different classes for each image. Though crucial to good performance on benchmarks, this post-processing is usually defined heuristically. We introduce a unified model for multi-class object recognition that casts the problem as a structured prediction task. Rather than predicting a binary label for each image window independently, our model simultaneously predicts a structured labeling of the entire image (Fig. 1). Our model learns statistics that capture the spatial arrangements of various object classes in real images, both in terms of which arrangements to suppress through NMS and which arrangements to favor through spatial co-occurrence statistics. We formulate parameter estimation in our model as a max-margin learning problem. Given training images with ground-truth object locations, we show how to formulate learning as a convex optimization problem. We employ the cutting plane algorithm of Joachims et al. (Mach. Learn. 2009) to efficiently learn a model from thousands

Abstract Many state-of-the-art approaches for object recognition reduce the problem to a 0-1 classification task. This allows one to leverage sophisticated machine learning techniques for training classifiers from labeled examples. However, these models are typically trained independently for each class using positive and negative examples cropped from images. At test-time, various post-processing heuristics such as non-maxima suppression (NMS) are required to reconcile multiple detections within and between different classes for each image. Though crucial to good performance on benchmarks, this post-processing is usually defined heuristically. We introduce a unified model for multi-class object recognition that casts the problem as a structured prediction task. Rather than predicting a binary label for each image window independently, our model simultaneously predicts a structured labeling of the entire image (Fig. 1). Our model learns statistics that capture the spatial arrangements of various object classes in real images, both in terms of which arrangements to suppress through NMS and which arrangements to favor through spatial co-occurrence statistics. We formulate parameter estimation in our model as a max-margin learning problem. Given training images with The Marr Prize is awarded to the best paper(s) at the biannual flagship vision conference, the IEEE International Conference on Computer Vision (ICCV). This paper is an extended and re-reviewed journal version of the 2009 prize-winning conference paper.

Most contemporary object detection approaches assume each object instance in the training data to be uniquely represented by a single bounding box. In this paper, we go beyond this conventional view by allowing an object instance to be described by multiple bounding boxes. The new bounding box annotations are determined based on the alignment of an object instance with the other training instances in the dataset. Our proposal enables the training data to be reused multiple times for training richer multi-component category models. We operationalize this idea by two complementary operations: bounding box shrinking, which finds subregions of an object instance that could be shared; and bounding box enlarging, which enlarges object instances to include local contextual cues. We empirically validate our approach on the PASCAL VOC detection dataset. 1

Most contemporary object detection approaches assume each object instance in the training data to be uniquely represented by a single bounding box. In this paper, we go beyond this conventional view by allowing an object instance to be described by multiple bounding boxes. The new bounding box annotations are determined based on the alignment of an object instance with the other training instances in the dataset. Our proposal enables the training data to be reused multiple times for training richer multi-component category models. We operationalize this idea by two complementary operations: bounding box shrinking, which finds subregions of an object instance that could be shared; and bounding box enlarging, which enlarges object instances to include local contextual cues. We empirically validate our approach on the PASCAL VOC detection dataset. 1