CVPR 2015 论文列表

In this paper, we introduce a spherical embedding technique to position a given set of silhouettes of an object as observed from a set of cameras arbitrarily positioned around the object. Our technique estimates dissimilarities among the silhouettes and embeds them directly in the rotation space SO(3). The embedding is obtained by an optimization scheme applied over the rotations represented with exponential maps. Since the measure for inter-silhouette dissimilarities contains many outliers, our key idea is to perform the embedding by only using a subset of the estimated dissimilarities. We present a technique that carefully screens for inlier-distances, and the pairwise scaled dissimilarities are embedded in a spherical space, diffeomorphic to SO(3). We show that our method outperforms spherical MDS embedding, demonstrate its performance on various multi-view sets, and highlight its robustness to outliers.

The fact that image data samples lie on a manifold has been successfully exploited in many learning and inference problems. In this paper we leverage the specific structure of data in order to improve recognition accuracies in general recognition tasks. In particular we propose a novel framework that allows to embed manifold priors into sparse representation-based classification (SRC) approaches. We also show that manifold constraints can be transferred from the data to the optimized variables if these are linearly correlated. Using this new insight, we define an efficient alternating direction method of multipliers (ADMM) that can consistently integrate the manifold constraints during the optimization process. This is based on the property that we can recast the problem as the projection over the manifold via a linear embedding method based on the Geodesic distance. The proposed approach is successfully applied on face, digit, action and objects recognition showing a consistently increase on performance when compared to the state of the art.

This paper addresses the problem of learning long binary codes from high-dimensional data. We observe that two key challenges arise while learning and using long binary codes: (1) lack of an effective regularizer for the learned high-dimensional mapping and (2) high computational cost for computing long codes. In this paper, we overcome both these problems by introducing a sparsity encouraging regularizer that reduces the effective number of parameters involved in the learned projection operator. This regularizer not only reduces overfitting but, due to the sparse nature of the projection matrix, also leads to a dramatic reduction in the computational cost. To evaluate the effectiveness of our method, we analyze its performance on the problems of nearest neighbour search, image retrieval and image classification. Experiments on a number of challenging datasets show that our method leads to better accuracy than dense projections (ITQ and LSH) with the same code lengths, and meanwhile is over an order of magnitude faster. Furthermore, our method is also more accurate and faster than other recently proposed methods for speeding up high-dimensional binary encoding.

We describe a method to produce detailed high resolution depth maps from aggressively subsampled depth measurements. Our method fully uses the relationship between image segmentation boundaries and depth boundaries. It uses an image combined with a low resolution depth map. 1) The image is segmented with the guidance of sparse depth samples. 2) Each segment has its depth field reconstructed independently using a novel smoothing method. 3) For videos, time-stamped samples from near frames are incorporated. The paper shows reconstruction results of super resolution from x4 to x100, while previous methods mainly work on x2 to x16. The method is tested on four different datasets and six video sequences, covering quite different regimes, and it outperforms recent state of the art methods quantitatively and qualitatively. We also demonstrate that depth maps produced by our method can be used by applications such as hand trackers, while depth maps from other methods have problems.

Deep neural networks have achieved remarkable performance in both image classification and object detection problems, at the cost of a large number of parameters and computational complexity. In this work, we show how to reduce the redundancy in these parameters using a sparse decomposition. Maximum sparsity is obtained by exploiting both inter-channel and intra-channel redundancy, with a fine-tuning step that minimize the recognition loss caused by maximizing sparsity. This procedure zeros out more than 90\% of parameters, with a drop of accuracy that is less than 1\% on the ILSVRC2012 dataset. We also propose an efficient sparse matrix multiplication algorithm on CPU for Sparse Convolutional Neural Networks (SCNN) models. Our CPU implementation demonstrates much higher efficiency than the off-the-shelf sparse matrix libraries, with a significant speedup realized over the original dense network. In addition, we apply the SCNN model to the object detection problem, in conjunction with a cascade model and sparse fully connected layers, to achieve significant speedups.

The quantization techniques have shown competitive performance in approximate nearest neighbor search. The state-of-the-art algorithm, composite quantization, takes advantage of the compositionabity, i.e., the vector approximation accuracy, as opposed to product quantization and Cartesian k-means. However, we have observed that the runtime cost of computing the distance table in composite quantization, which is used as a lookup table for fast distance computation, becomes nonnegligible in real applications, e.g., reordering the candidates retrieved from the inverted index when handling very large scale databases. To address this problem, we develop a novel approach, called sparse composite quantization, which constructs sparse dictionaries. The benefit is that the distance evaluation between the query and the dictionary element (a sparse vector) is accelerated using the efficient sparse vector operation, and thus the cost of distance table computation is reduced a lot. Experiment results on large scale ANN retrieval tasks (1M SIFTs and 1B SIFTs) and applications to object retrieval show that the proposed approach yields competitive performance: superior search accuracy to product quantization and Cartesian k-means with almost the same computing cost, and much faster ANN search than composite quantization with the same level of accuracy.

Human actions are, inherently, structured patterns of body movements. We explore ensembles of hierarchical spatio-temporal trees, discovered directly from training data, to model these structures for action recognition. The hierarchical spatio-temporal trees provide a robust mid-level representation for actions. However, discovery of frequent and discriminative tree structures is challenging due to the exponential search space, particularly if one allows partial matching. We address this by first building a concise action vocabulary via discriminative clustering. Using the action vocabulary we then utilize tree mining with subsequent tree clustering and ranking to select a compact set of highly discriminative tree patterns. We show that these tree patterns, alone, or in combination with shorter patterns (action words and pair-wise patterns) achieve state-of-the-art performance on two challenging datasets: UCF Sports and HighFive. Moreover, trees learned on HighFive are used in recognizing two action classes in a different dataset, Hollywood3D, demonstrating the potential for cross-dataset generality of the trees our approach discovers.

Jigsaw-puzzle solving is necessary in many applications, including biology, archaeology, and every-day life. In this paper we consider the square jigsaw puzzle problem, where the goal is to reconstruct the image from a set of non-overlapping, unordered, square puzzle parts. Our key contribution is a fast, fully-automatic, and general solver, which assumes no prior knowledge about the original image. It is general in the sense that it can handle puzzles of unknown size, with pieces of unknown orientation, and even puzzles with missing pieces. Moreover, it can handle all the above, given pieces from multiple puzzles. Through an extensive evaluation we show that our approach outperforms state-of-the-art methods on commonly-used datasets.

This paper presents a method to predict social saliency, the likelihood of joint attention, given an input image or video by leveraging the social interaction data captured by first person cameras. Inspired by electric dipole moments, we introduce a social formation feature that encodes the geometric relationship between joint attention and its social formation. We learn this feature from the first person social interaction data where we can precisely measure the locations of joint attention and its associated members in 3D. An ensemble classifier is trained to learn the geometric relationship. Using the trained classifier, we predict social saliency in real-world scenes with multiple social groups including scenes from team sports captured in a third person view. Our representation does not require directional measurements such as gaze directions. A geometric analysis of social interactions in terms of the F-formation theory is also presented.

Structural regularities in man-made environments reflect in the distribution of their surface normals. Describing these surface normal distributions is important in many computer vision applications, such as scene understanding, plane segmentation, and regularization of 3D reconstructions. Based on the small-variance limit of Bayesian nonparametric von-Mises-Fisher (vMF) mixture distributions, we propose two new flexible and efficient k-means-like clustering algorithms for directional data such as surface normals. The first, DP-vMF-means, is a batch clustering algorithm derived from the Dirichlet process (DP) vMF mixture. Recognizing the sequential nature of data collection in many applications, we extend this algorithm to DDP-vMF-means, which infers temporally evolving cluster structure from streaming data. Both algorithms naturally respect the geometry of directional data, which lies on the unit sphere. We demonstrate their performance on synthetic directional data and real 3D surface normals from RGB-D sensors. While our experiments focus on 3D data, both algorithms generalize to high dimensional directional data such as protein backbone configurations and semantic word vectors.

We propose a novel object detection framework for partially-occluded small instances, such as pedestrians in low resolution surveillance video, cells under a microscope, flocks of small animals (e.g. birds, fishes), or even tiny insects like honeybees and flies. These scenarios are very challenging for traditional detectors, which are typically trained on individual instances. In our approach, we first estimate the object density map of the input image, and then divide it into local regions. For each region, a sliding window (ROI) is passed over the density map to calculate the instance count within each ROI. 2D integer programming is used to recover the locations of object instances from the set of ROI counts, and the global count estimate of the density map is used as a constraint to regularize the detection performance. Finally, the bounding box for each instance is estimated using the local density map. Compared with current small-instance detection methods, our proposed approach achieves state-of-the-art performance on several challenging datasets including fluorescence microscopy cell images, UCSD pedestrians, small animals and insects.

Retrieving 3D models from 2D human sketches has received considerable attention in the areas of graphics, image retrieval, and computer vision. Almost always in state of the art approaches a large amount of ``best views'' are computed for 3D models, with the hope that the query sketch matches one of these 2D projections of 3D models using predefined features. We argue that this two stage approach (view selection -- matching) is pragmatic but also problematic because the ``best views'' are subjective and ambiguous, which makes the matching inputs obscure. This imprecise nature of matching further makes it challenging to choose features manually. Instead of relying on the elusive concept of ``best views'' and the hand-crafted features, we propose to define our views using a minimalism approach and learn features for both sketches and views. Specifically, we drastically reduce the number of views to only two predefined directions for the whole dataset. Then, we learn two Siamese Convolutional Neural Networks (CNNs), one for the views and one for the sketches. The loss function is defined on the within-domain as well as the cross-domain similarities. Our experiments on three benchmark datasets demonstrate that our method is significantly better than state of the art approaches, and outperforms them in all conventional metrics.

Intuitively, the appearance of true object boundaries varies from image to image. Hence the usual monolithic approach of training a single boundary predictor and applying it to all images regardless of their content is bound to be suboptimal. In this paper we therefore propose Situational Object Boundary Detection: we first define a variety of situations and train a specialized object boundary detector for each of them using [Dollar13]. Then given a test image, we classify it into these situations using its context, which we model by global image appearance. We apply the corresponding situational object boundary detectors, and fuse them based on the classification probabilities. In experiments on ImageNet [Russakovsky14], Microsoft COCO [Lin], and Pascal VOC 2012 segmentation [Everingham14] we show that our situational object boundary detection gives significant improvements over a monolithic approach. Additionally, our method substantially outperforms [Hariharan11iccv] on semantic contour detection on their SBD dataset.

The camera response function (CRF) relates quantised image pixel values with physical incoming light. This paper describes a method to estimate the CRF from a single image of a general two-coloured surface for which the albedo ratio between the coloured regions is known a priori. While other radiometric calibration methods either use multiple frames or require the light to be infinitely distant, the algorithm herein proposed makes no assumptions about lighting conditions and can handle cameras with strong vignetting. Although the approach is generic, in the sense that can be applied to any camera system, the method is particularly well suited for determining the CRF of near-lighting endoscopes in the operating room. This is a very pertinent problem for which no practical, effective solutions have been proposed. The robustness, repeatability, and accuracy of the algorithm is experimentally validated in real images acquired with different endoscopic set-ups.

This paper presents a method for single target tracking of arbitrary objects in challenging video sequences. Targets are modeled at three different levels of granularity (pixel level, parts-based level and bounding box level), which are cross-constrained to enable robust model relearning. The main contribution is an adaptive clustered decision tree method which dynamically selects the minimum combination of features necessary to sufficiently represent each target part at each frame, thereby providing robustness with computational efficiency. The adaptive clustered decision tree is implemented in two separate parts of the tracking algorithm: firstly to enable robust matching at the parts-based level between successive frames; and secondly to select the best superpixels for learning new parts of the target. We have tested the tracker using two different tracking benchmarks (VOT2013-2014 and CVPR2013 tracking challenges), based on two different test methodologies, and show it to be significantly more robust than the best state-of-the-art methods from both of those tracking challenges, while also offering competitive tracking precision.

Self-similarity based super-resolution (SR) algorithms are able to produce visually pleasing results without extensive training on external databases. Such algorithms exploit the statistical prior that patches in a natural image tend to recur within and across scales of the same image. However, the internal dictionary obtained from the given image may not always be sufficiently expressive to cover the textural appearance variations in the scene. In this paper, we extend self-similarity based SR to overcome this drawback. We expand the internal patch search space by allowing geometric variations. We do so by explicitly localizing planes in the scene and using the detected perspective geometry to guide the patch search process. We also incorporate additional affine transformations to accommodate local shape variations. We propose a compositional model to simultaneously handle both types of transformations. We extensively evaluate the performance in both urban and natural scenes. Even without using any external training databases, we achieve significantly superior results on urban scenes, while maintaining comparable performance on natural scenes as other state-of-the-art SR algorithms.

We present a method to jointly estimate scene depth and recover the clear latent image from a foggy video sequence. In our formulation, the depth cues from stereo matching and fog information reinforce each other, and produce superior results than conventional stereo or defogging algorithms. We first improve the photo-consistency term to explicitly model the appearance change due to the scattering effects. The prior matting Laplacian constraint on fog transmission imposes a detail-preserving smoothness constraint on the scene depth. We further enforce the ordering consistency between scene depth and fog transmission at neighboring points. These novel constraints are formulated together in an MRF framework, which is optimized iteratively by introducing auxiliary variables. The experiment results on real videos demonstrate the strength of our method.

We present a novel system which incorporates photometric stereo with the Time-of-Flight depth sensor. Adding to the classic ToF, the system utilizes multiple point light sources that enable the capturing of a normal field whilst taking depth images. Two calibration methods are proposed to determine the light sources' positions given the ToF sensor's relatively low resolution. An iterative refinement algorithm is formulated to account for the extra phase delays caused by the positions of the light sources. We find in experiments that the system is comparable to the classic ToF in depth accuracy, and it is able to recover finer details that are lost due to the noise level of the ToF sensor.

In this paper, we propose a sequential solution to simultaneously estimate camera pose and non-rigid 3D shape from a monocular video. In contrast to most existing approaches that rely on global representations of the shape, we model the object at a local level, as an ensemble of particles, each ruled by the linear equation of the Newton's second law of motion. This dynamic model is incorporated into a bundle adjustment framework, in combination with simple regularization components that ensure temporal and spatial consistency of the estimated shape and camera poses. The resulting approach is both efficient and robust to several artifacts such as noisy and missing data or sudden camera motions, while it does not require any training data at all. Validation is done in a variety of real video sequences, including articulated and non-rigid motion, both for continuous and discontinuous shapes. Our system is shown to perform comparable to competing batch, computationally expensive, methods and shows remarkable improvement with respect to the sequential ones.