CVPR 2016 Papers

Existing methods for pixel-wise labelling tasks generally disregard the underlying structure of labellings, often leading to predictions that are visually implausible. While incorporating structure into the model should improve prediction quality, doing so is challenging - manually specifying the form of structural constraints may be impractical and inference often becomes intractable even if structural constraints are given. We sidestep this problem by reducing structured prediction to a sequence of unconstrained prediction problems and demonstrate that this approach is capable of automatically discovering priors on shape, contiguity of region predictions and smoothness of region contours from data without any a priori specification. On the instance segmentation task, this method outperforms the state-of-the-art, achieving a mean AP^r of 63.6% at 50% overlap and 43.3% at 70% overlap.

Feature representations, both hand-designed and learned ones, are often hard to analyze and interpret, even when they are extracted from visual data. We propose a new approach to study image representations by inverting them with an up-convolutional neural network. We apply the method to shallow representations (HOG, SIFT, LBP), as well as to deep networks. For shallow representations our approach provides significantly better reconstructions than existing methods, revealing that there is surprisingly rich information contained in these features. Inverting a deep network trained on ImageNet provides several insights into the properties of the feature representation learned by the network. Most strikingly, the colors and the rough contours of an image can be reconstructed from activations in higher network layers and even from the predicted class probabilities.

An increasing number of computer vision tasks can be tackled with deep features, which are the intermediate outputs of a pre-trained Convolutional Neural Network. Despite the astonishing performance, deep features extracted from low-level neurons are still below satisfaction, arguably because they cannot access the spatial context contained in the higher layers. In this paper, we present InterActive, a novel algorithm which computes the activeness of neurons and network connections. Activeness is propagated through a neural network in a top-down manner, carrying high-level context and improving the descriptive power of low-level and mid-level neurons. Visualization indicates that neuron activeness can be interpreted as spatial-weighted neuron responses. We achieve state-of-the-art classification performance on a wide range of image datasets.

We address the problem of segmenting multiple object instances in complex videos. Our method does not require manual pixel-level annotation for training, and relies instead on readily-available object detectors or visual object tracking only. Given object bounding boxes at input, we cast video segmentation as a weakly-supervised learning problem. Our proposed objective combines (a) a discriminative clustering term for background segmentation, (b) a spectral clustering one for grouping pixels of same object instances, and (c) linear constraints enabling instance-level segmentation. We propose a convex relaxation of this problem and solve it efficiently using the Frank-Wolfe algorithm. We report results and compare our method to several baselines on a new video dataset for multi-instance person segmentation.

Our aim is to provide a pixel-wise instance-level labeling of a monocular image in the context of autonomous driving. We build on recent work [Zhang et al., ICCV15] that trained a convolutional neural net to predict instance labeling in local image patches, extracted exhaustively in a stride from an image. A simple Markov random field model using several heuristics was then proposed in [Zhang et al., ICCV15] to derive a globally consistent instance labeling of the image. In this paper, we formulate the global labeling problem with a novel densely connected Markov random field and show how to encode various intuitive potentials in a way that is amenable to efficient mean field inference [Krahenbuhl et al., NIPS11]. Our potentials encode the compatibility between the global labeling and the patch-level predictions, contrast-sensitive smoothness as well as the fact that separate regions form different instances. Our experiments on the challenging KITTI benchmark [Geiger et al., CVPR12] demonstrate that our method achieves a significant performance boost over the baseline [Zhang et al., ICCV15].

It is well known that contextual and multi-scale representations are important for accurate visual recognition. In this paper we present the Inside-Outside Net (ION), an object detector that exploits information both inside and outside the region of interest. Contextual information outside the region of interest is integrated using spatial recurrent neural networks. Inside, we use skip pooling to extract information at multiple scales and levels of abstraction. Through extensive experiments we evaluate the design space and provide readers with an overview of what tricks of the trade are important. ION improves state-of-the-art on PASCAL VOC 2012 object detection from 73.9% to 77.9% mAP. On the new and more challenging MS COCO dataset, we improve state-of-the-art from 19.7% to 33.1% mAP. In the 2015 MS COCO Detection Challenge, our ION model won "Best Student Entry" and finished 3rd place overall. As intuition suggests, our detection results provide strong evidence that context and multi-scale representations improve small object detection.

We explore the visual recognition problem from a main data view when an auxiliary data view is available during training. This is important because it allows improving the training of visual classifiers when paired additional data is cheaply available, and it improves the recognition from multi-view data when there is a missing view at testing time. The problem is challenging because of the intrinsic asymmetry caused by the missing auxiliary view during testing. We account for such view during training by extending the information bottleneck method, and by combining it with risk minimization. In this way, we establish an information theoretic principle for leaning any type of visual classifier under this particular setting. We use this principle to design a large-margin classifier with an efficient optimization in the primal space. We extensively compare our method with the state-of-the-art on different visual recognition datasets, and with different types of auxiliary data, and show that the proposed framework has a very promising potential.

We present a global and convex formulation for template-less 3D reconstruction of a deforming object with the perspective camera. We show for the first time how to construct a Second-Order Cone Programming (SOCP) problem for Non-Rigid Shape-from-Motion (NRSfM) using the Maximum-Depth Heuristic (MDH). In this regard, we deviate strongly from the general trend of using affine cameras and factorization-based methods to solve NRSfM. In MDH, the points' depths are maximized so that the distance between neighbouring points in camera space are upper bounded by the geodesic distance. In NRSfM both geodesic and camera space distances are unknown. We show that, nonetheless, given point correspondences and the camera's intrinsics the whole problem is convex and solvable with SOCP. We show with extensive experiments that our method accurately reconstructs quasi-isometric surfaces from partial views under articulated and strong deformations. It naturally handles missing correspondences, non-smooth objects and is very simple to implement compared to previous methods, with only one free parameter (the neighbourhood size).

Abstract In this paper, we address the problem of object discovery in time-varying, large-scale image collections. A core part of our approach is a novel Limited Horizon Minimum Spanning Tree (LH-MST) structure that closely approximates the Minimum Spanning Tree at a small fraction of the latter's computational cost. Our proposed tree structure can be created in a local neighborhood of the matching graph during image retrieval and can be efficiently updated whenever the image database is extended. We show how the LH-MST can be used within both single-link hierarchical agglomer- ative clustering and the Iconoid Shift framework for object discovery in image collections, resulting in significant efficiency gains and making both approaches capable of incremental clustering with online updates. We evaluate our approach on a dataset of 500k images from the city of Paris and compare its results to the batch version of both clustering algorithms.

We present a new algorithm for multi-region segmentation of 2D images with objects that may partially occlude each other. Our algorithm is based on the observation that human performance on this task is based both on prior knowledge about plausible shapes and taking into account the presence of occluding objects whose shape is already known - once an occluded region is identified, the shape prior can be used to guess the shape of the missing part. We capture the former aspect using a deep learning model of shape; for the latter, we simultaneously minimize the energy of all regions and consider only unoccluded pixels for data agreement. Existing algorithms incorporating object shape priors consider every object separately in turn and can't distinguish genuine deviation from the expected shape from parts missing due to occlusion. We show that our method significantly improves on the performance of a representative algorithm, as evaluated on both preprocessed natural and synthetic images. Furthermore, on the synthetic images, we recover the ground truth segmentation with good accuracy.

In this paper we address the issue of output instability of deep neural networks: small perturbations in the visual input can significantly distort the feature embeddings and output of a neural network. Such instability affects many deep architectures with state-of-the-art performance on a wide range of computer vision tasks. We present a general stability training method to stabilize deep networks against small input distortions that result from various types of common image processing, such as compression, rescaling, and cropping. We validate our method by stabilizing the state-of-the-art Inception architecture against these types of distortions. In addition, we demonstrate that our stabilized model gives robust state-of-the-art performance on large-scale near-duplicate detection, similar-image ranking, and classification on noisy datasets.

Person re-identification is the problem of recognizing people across images or videos from non-overlapping views. Although there has been much progress in person re-identification for the last decade, it still remains a challenging task because of severe appearance changes of a person due to diverse camera viewpoints and person poses. In this paper, we propose a novel framework for person re-identification by analyzing camera viewpoints and person poses, so-called Pose-aware Multi-shot Matching (PaMM), which robustly estimates target poses and efficiently conducts multi-shot matching based on the target pose information. Experimental results using public person re-identification dataset show that the proposed methods are promising for person re-identification under diverse viewpoints and pose variances.

In this paper we consider the task of recognizing human actions in realistic video where human actions are dominated by irrelevant factors. We first study the benefits of removing non-action video segments, which are the ones that do not portray any human action. We then learn a non-action classifier and use it to down-weight irrelevant video segments. The non-action classifier is trained using ActionThread, a dataset with shot-level annotation for the occurrence or absence of a human action. The non-action classifier can be used to identify non-action shots with high precision and subsequently used to improve the performance of action recognition systems.

This paper addresses the problem of ultra-large-scale search in Hamming spaces. There has been considerable research on generating compact binary codes in vision, for example for visual search tasks. However the issue of efficient searching through huge sets of binary codes remains largely unsolved. To this end, we propose a novel, unsupervised approach to thresholded search in Hamming space, supporting long codes (e.g. 512-bits) with a wide-range of Hamming distance radii. Our method is capable of working efficiently with billions of codes delivering between one to three orders of magnitude acceleration, as compared to prior art. This is achieved by relaxing the equal-size constraint in the Multi-Index Hashing approach, leading to multiple hash-tables with variable length hash-keys. Based on the theoretical analysis of the retrieval probabilities of multiple hash-tables we propose a novel search algorithm for obtaining a suitable set of hash-key lengths. The resulting retrieval mechanism is shown empirically to improve the efficiency over the state-of-the-art, across a range of datasets, bit-depths and retrieval thresholds.

Removing image blur caused by camera shake is an ill-posed problem, as both the latent image and the point spread function (PSF) are unknown. A recent approach to address this problem is to record camera motion through inertial sensors, i.e., gyroscopes and accelerometers, and then reconstruct spatially-variant PSFs from these readings. While this approach has been effective for high-quality inertial sensors, it has been infeasible for the inertial sensors in smartphones, which are of relatively low quality and present a number of challenging issues, including varying sensor parameters, high sensor noise, and calibration error. In this paper, we identify the issues that plague smartphone inertial sensors and propose a solution that successfully utilizes the sensor readings for image deblurring. With both the sensor data and the image itself, the proposed method is able to accurately estimate the sensor parameters online and also the spatially-variant PSFs for enhanced deblurring performance. The effectiveness of this technique is demonstrated in experiments on a popular mobile phone. With this approach, the quality of image deblurring can be appreciably raised on the most common of imaging devices.

This paper proposes a novel technique to extract training data from free space in a scene using a stereo camera. The proposed technique exploits the projection of planes in the v-disparity image paired with Bayesian linear regression to reliably identify training image pixels belonging to free space in a scene. Unlike other methods in the literature, the algorithm does not require any prior training, has only one free parameter, and is shown to provide consistent results over a variety of terrains without the need for any manual tuning. The proposed method is compared to two other data extraction methods from the literature. Results of Support Vector classifiers using training data extracted by the proposed technique are superior in terms of quality and consistency of free space estimation. Furthermore, the computation time required by the proposed technique is shown to be smaller and more consistent than that of other training data extraction methods.

We address the problem of estimating image difficulty defined as the human response time for solving a visual search task. We collect human annotations of image difficulty for the PASCAL VOC 2012 data set through a crowd-sourcing platform. We then analyze what human interpretable image properties can have an impact on visual search difficulty, and how accurate are those properties for predicting difficulty. Next, we build a regression model based on deep features learned with state of the art convolutional neural networks and show better results for predicting the ground-truth visual search difficulty scores produced by human annotators. Our model is able to correctly rank about 75% image pairs according to their difficulty score. We also show that our difficulty predictor generalizes well to new classes not seen during training. Finally, we demonstrate that our predicted difficulty scores are useful for weakly supervised object localization (8% improvement) and semi-supervised object classification (1% improvement).

Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods and the "perfect single frame detector". We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech dataset), and by manually clustering the recurrent errors of a top detector. Our results characterise both localisation and background-versus-foreground errors. To address localisation errors we study the impact of training annotation noise on the detector performance, and show that we can improve even with a small portion of sanitised training data. To address background/foreground discrimination, we study convnets for pedestrian detection, and discuss which factors affect their performance. Other than our in-depth analysis, we report top performance on the Caltech dataset, and provide a new sanitised set of training and test annotations.

How to avoid ambiguity is a challenging problem for conic-based homography estimation. In this paper, we address the problem of homography estimation from two separate ellipses. We find that any two ellipses have a unique common self-polar triangle, which can provide three line correspondences. Furthermore, by investigating the location features of the common self-polar triangle, we show that one vertex of the triangle lies outside of both ellipses, while the other two vertices lies inside the ellipses separately. Accordingly, one more line correspondence can be obtained from the intersections of the conics and the common self-polar triangle. Therefore, four line correspondences can be obtained based on the common self-polar triangle, which can provide enough constraints for the homography estimation. The main contributions in this paper include: (1) A new discovery on the location features of the common self-polar triangle of separate ellipses. (2) A novel approach for homography estimation. Simulate experiments and real experiments are conducted to demonstrate the feasibility and accuracy of our approach.