CVPR 2016 Papers

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.

Rendering the semantic content of an image in different styles is a difficult image processing task. Arguably, a major limiting factor for previous approaches has been the lack of image representations that explicitly represent semantic information and, thus, allow to separate image content from style. Here we use image representations derived from Convolutional Neural Networks optimised for object recognition, which make high level image information explicit. We introduce A Neural Algorithm of Artistic Style that can separate and recombine the image content and style of natural images. The algorithm allows us to produce new images of high perceptual quality that combine the content of an arbitrary photograph with the appearance of numerous well-known artworks. Our results provide new insights into the deep image representations learned by Convolutional Neural Networks and demonstrate their potential for high level image synthesis and manipulation.

We tackle image question answering (ImageQA) problem by learning a convolutional neural network (CNN) with a dynamic parameter layer whose weights are determined adaptively based on questions. For the adaptive parameter prediction, we employ a separate parameter prediction network, which consists of gated recurrent unit (GRU) taking a question as its input and a fully-connected layer generating a set of candidate weights as its output. However, it is challenging to construct a parameter prediction network for a large number of parameters in the fully-connected dynamic parameter layer of the CNN. We reduce the complexity of this problem by incorporating a hashing technique, where the candidate weights given by the parameter prediction network are selected using a predefined hash function to determine individual weights in the dynamic parameter layer. The proposed network---joint network with the CNN for ImageQA and the parameter prediction network---is trained end-to-end through back-propagation, where its weights are initialized using a pre-trained CNN and GRU. The proposed algorithm illustrates the state-of-the-art performance on all available public ImageQA benchmarks.

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.

Automatically generating a natural language description of an image has attracted interests recently both because of its importance in practical applications and because it connects two major artificial intelligence fields: computer vision and natural language processing. Existing approaches are either top-down, which start from a gist of an image and convert it into words, or bottom-up, which come up with words describing various aspects of an image and then combine them. In this paper, we propose a new algorithm that combines both approaches through a model of semantic attention. Our algorithm learns to selectively attend to semantic concept proposals and fuse them into hidden states and outputs of recurrent neural networks. The selection and fusion form a feedback connecting the top-down and bottom-up computation. We evaluate our algorithm on two public benchmarks: Microsoft COCO and Flickr30K. Experimental results show that our algorithm significantly outperforms the state-of-the-art approaches consistently across different evaluation metrics.

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.

Almost all of the current top-performing object detection networks employ region proposals to guide the search for object instances. State-of-the-art region proposal methods usually need several thousand proposals to get high recall, thus hurting the detection efficiency. Although the latest Region Proposal Network method gets promising detection accuracy with several hundred proposals, it still struggles in small-size object detection and precise localization (e.g., large IoU thresholds), mainly due to the coarseness of its feature maps. In this paper, we present a deep hierarchical network, namely HyperNet, for handling region proposal generation and object detection jointly. Our HyperNet is primarily based on an elaborately designed Hyper Feature which aggregates hierarchical feature maps first and then compresses them into a uniform space. The Hyper Features well incorporate deep but highly semantic, intermediate but really complementary, and shallow but naturally high-resolution features of the image, thus enabling us to construct HyperNet by sharing them both in generating proposals and detecting objects via an end-to-end joint training strategy. For the deep VGG16 model, our method achieves completely leading recall and state-of-the-art object detection accuracy on PASCAL VOC 2007 and 2012 using only 100 proposals per image. It runs with a speed of 5 fps (including all steps) on a GPU, thus having the potential for real-time processing.

Structured light sensors are popular due to their robustness to untextured scenes and multipath. These systems triangulate depth by solving a correspondence problem between each camera and projector pixel. This is often framed as a local stereo matching task, correlating patches of pixels in the observed and reference image. However, this is computationally intensive, leading to reduced depth accuracy and framerate. We contribute an algorithm for solving this correspondence problem efficiently, without compromising depth accuracy. For the first time, this problem is cast as a classification-regression task, which we solve extremely efficiently using an ensemble of cascaded random forests. Our algorithm scales in number of disparities, and each pixel can be processed independently, and in parallel. No matching or even access to the corresponding reference pattern is required at runtime, and regressed labels are directly mapped to depth. Our GPU-based algorithm runs at a 1KHz for 1.3MP input/output images, with disparity error of 0.1 subpixels. We show a prototype high framerate depth camera running at 375Hz, useful for solving tracking-related problems. We demonstrate our algorithmic performance, creating high resolution real-time depth maps that surpass the quality of current state of the art depth technologies, highlighting quantization-free results with reduced holes, edge fattening and other stereo-based depth artifacts.

Hierarchical feature extractors such as Convolutional Networks (ConvNets) have achieved impressive performance on a variety of classification tasks using purely feedforward processing. Feedforward architectures can learn rich representations of the input space but do not explicitly model dependencies in the output spaces, that are quite structured for tasks such as articulated human pose estimation or object segmentation. Here we propose a framework that expands the expressive power of hierarchical feature extractors to encompass both input and output spaces, by introducing top-down feedback. Instead of directly predicting the outputs in one go, we use a self-correcting model that progressively changes an initial solution by feeding back error predictions, in a process we call Iterative Error Feedback (IEF). IEF shows excellent performance on the task of articulated pose estimation in the challenging MPII and LSP benchmarks, matching the state-of-the-art without requiring ground truth scale annotation.

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.

This paper presents a highway vehicle counting method in compressed domain, aiming at achieving acceptable estimation performance approaching the pixel-domain methods. Such a task essentially is challenging because the available information (e.g. motion vector) to describe vehicles in videos is quite limited and inaccurate, and the vehicle count in realistic traffic scenes always varies greatly. To tackle this issue, we first develop a batch of low-level features, which can be extracted from the encoding metadata of videos, to mitigate the informational insufficiency of compressed videos. Then we propose a Hierarchical Classification based Regression (HCR) model to estimate the vehicle count from features. HCR hierarchically divides the traffic scenes into different cases according to vehicle density, such that the broad-variation characteristics of traffic scenes can be better approximated. Finally, we evaluated the proposed method on the real highway surveillance videos. The results show that our method is very competitive to the pixel-domain methods, which can reach similar performance along with its lower complexity.

The emergence of wearable devices such as portable cameras and smart glasses makes it possible to record life logging first-person videos. Browsing such long unstructured videos is time-consuming and tedious. This paper studies the discovery of moments of user's major or special interest (i.e., highlights) in a video, for generating the summarization of first-person videos. Specifically, we propose a novel pairwise deep ranking model that employs deep learning techniques to learn the relationship between highlight and non-highlight video segments. A two-stream network structure by representing video segments from complementary information on appearance of video frames and temporal dynamics across frames is developed for video highlight detection. Given a long personal video, equipped with the highlight detection model, a highlight score is assigned to each segment. The obtained highlight segments are applied for summarization in two ways: video timelapse and video skimming. The former plays the highlight (non-highlight) segments at low (high) speed rates, while the latter assembles the sequence of segments with the highest scores. On 100 hours of first-person videos for 15 unique sports categories, our highlight detection achieves the improvement over the state-of-the-art RankSVM method by 10.5% in terms of accuracy. Moreover, our approaches produce video summary with better quality by a user study from 35 human subjects.

We propose a novel approach that generates a high-quality depth map from a set of images captured with a small viewpoint variation, namely small motion clip. As opposed to prior methods that recover scene geometry and camera motions using pre-calibrated cameras, we introduce a self-calibrating bundle adjustment tailored for small motion. This allows our dense stereo algorithm to produce a high-quality depth map for the user without the need for camera calibration. In the dense matching, the distributions of intensity profiles are analyzed to leverage the benefit of having negligible intensity changes within the scene due to the minuscule variation in viewpoint. The depth maps obtained by the proposed framework show accurate and extremely fine structures that are unmatched by previous literature under the same small motion configuration.

Semantic segmentation aims to parse the scene structure of images by annotating the labels to each pixel so that images can be segmented into different regions. While image structures usually have various scales, it is difficult to use a single scale to model the spatial contexts for all individual pixels. Multi-scale Convolutional Neural Networks (CNNs) and their variants have made striking success for modeling the global scene structure for an image. However, they are limited in labeling fine-grained local structures like pixels and patches, since spatial contexts might be blindly mixed up without appropriately customizing their scales. To address this challenge, we develop a novel paradigm of multi-scale deep network to model spatial contexts surrounding different pixels at various scales. It builds multiple layers of memory cells, learning feature representations for individual pixels at their customized scales by hierarchically absorbing relevant spatial contexts via memory gates between layers.Such Hierarchically Gated Deep Networks (HGDNs) can customize a suitable scale for each pixel, thereby delivering better performance on labeling scene structures of various scales. We conduct the experiments on two datasets, and show competitive results compared with the other multi-scale deep networks on the semantic segmentation task.

Recently, deep learning approach, especially deep Convolutional Neural Networks (ConvNets), have achieved overwhelming accuracy with fast processing speed for image classification. Incorporating temporal structure with deep ConvNets for video representation becomes a fundamental problem for video content analysis. In this paper, we propose a new approach, namely Hierarchical Recurrent Neural Encoder (HRNE), to exploit temporal information of videos. Compared to recent video representation inference approaches, this paper makes the following three contributions. First, our HRNE is able to efficiently exploit video temporal structure in a longer range by reducing the length of input information flow, and compositing multiple consecutive inputs at a higher level. Second, computation operations are significantly lessened while attaining more non-linearity. Third, HRNE is able to uncover temporal transitions between frame chunks with different granularities, i.e. it can model the temporal transitions between frames as well as the transitions between segments. We apply the new method to video captioning where temporal information plays a crucial role. Experiments demonstrate that our method outperforms the state-of-the-art on video captioning benchmarks.