Showing papers by "Anurag Mittal published in 2015"

Co-operative Pedestrians Group Tracking in Crowded Scenes Using an MST Approach

Abstract: We address the problem of multiple pedestrian tracking in crowded scenes in videos recorded by a static uncalibrated camera. We propose an online multiple pedestrian tracking algorithm that utilizes group behaviour of pedestrians using minimum spanning trees (MST). We first divide pedestrians into several groups using the agglomerative hierarchical clustering, taking position and velocity of pedestrians as features, and then we track each group, represented by an MST, with the pictorial structures method. We also propose a method to detect and handle interpedestrian occlusions using a custom trained head detector for crowded scenes. Finally, we present experiments on two challenging and publicly available datasets and show improvements on multiple object tracking accuracy (MOTA) over other methods.

Sketch-based Image Retrieval from Millions of Images under Rotation, Translation and Scale Variations.

Abstract: Proliferation of touch-based devices has made sketch-based image retrieval practical. While many methods exist for sketch-based object detection/image retrieval on small datasets, relatively less work has been done on large (web)-scale image retrieval. In this paper, we present an efficient approach for image retrieval from millions of images based on user-drawn sketches. Unlike existing methods for this problem which are sensitive to even translation or scale variations, our method handles rotation, translation, scale (i.e. a similarity transformation) and small deformations. The object boundaries are represented as chains of connected segments and the database images are pre-processed to obtain such chains that have a high chance of containing the object. This is accomplished using two approaches in this work: a) extracting long chains in contour segment networks and b) extracting boundaries of segmented object proposals. These chains are then represented by similarity-invariant variable length descriptors. Descriptor similarities are computed by a fast Dynamic Programming-based partial matching algorithm. This matching mechanism is used to generate a hierarchical k-medoids based indexing structure for the extracted chains of all database images in an offline process which is used to efficiently retrieve a small set of possible matched images for query chains. Finally, a geometric verification step is employed to test geometric consistency of multiple chain matches to improve results. Qualitative and quantitative results clearly demonstrate superiority of the approach over existing methods.

Mixture of parts revisited: Expressive part interactions for Pose Estimation

Abstract: Part-based models with restrictive tree-structured interactions for the Human Pose Estimation problem, leave many part interactions unhandled. Two of the most common and strong manifestations of such unhandled interactions are self-occlusion among the parts and the confusion in the localization of the non-adjacent symmetric parts. By handling the self-occlusion in a data efficient manner, we improve the performance of the basic Mixture of Parts model by a large margin, especially on difficult poses. We address the confusion in the symmetric limb localization using a combination of two complementing trees, showing an improvement in the performance on all the parts with a very small trade-off in the running time. Finally, we show that the combination of the two solutions improves the results. We compare our HOG-based method with other methods using similar features and report results equivalent to the best method on two standard datasets with a large reduction in the running time.

Adaptive Locally Affine-Invariant Shape Matching

Abstract: Matching deformable objects using their shapes is an important problem in computer vision since shape is perhaps the most distinguishable characteristic of an object. The problem is difficult due to many factors such as intra-class variations, local deformations, articulations, viewpoint changes and missed and extraneous contour portions due to errors in shape extraction. While small local deformations has been handled in the literature by allowing some leeway in the matching of individual contour points via methods such as Chamfer distance and Hausdorff distance, handling more severe deformations and articulations has been done by applying local geometric corrections such as similarity or affine. However, determining which portions of the shape should be used for the geometric corrections is very hard, although some methods have been tried. In this paper, we address this problem by an efficient search for the group of contour segments to be clustered together for a geometric correction using Dynamic Programming by essentially searching for the segmentations of two shapes that lead to the best matching between them. At the same time, we allow portions of the contours to remain unmatched to handle missing and extraneous contour portions. Experiments indicate that our method outperforms other algorithms, especially when the shapes to be matched are more complex.

Revisiting pose estimation with foreshortening compensation and color information

Abstract: This paper addresses the problem of upper body pose estimation. The task is to detect and estimate 2D human configuration in static images for six parts: head, torso, and left-right upper and lower arms. The common approach to solve this has been the Pictorial Structure method (Felzenszwalb and Huttenlocher, 2005). We present this as a graphical model inference problem and use the loopy belief propagation algorithm for inference. When a human appears in fronto-parallel plane, fixed size part detectors are sufficient and give reliable detection. But when parts like lower and upper arms move out of the plane, we observe foreshortening and the part detectors become erroneous. We propose an approach that compensates foreshortening in the upper and lower arms, and effectively prunes the search state space of each part. Additionally, we introduce two extra pairwise constraints to exploit the color similarity information between parts during inference to get better localization of the upper and lower arms. Finally, we present experiments and results on two challenging datasets (Buffy and ETHZ Pascal), showing improvements on the lower arms accuracy and comparable results for other parts.