CVPR 2016 论文列表

Projective analysis is an important solution for 3D shape retrieval, since human visual perceptions of 3D shapes rely on various 2D observations from different view points. Although multiple informative and discriminative views are utilized, most projection-based retrieval systems suffer from heavy computational cost, thus cannot satisfy the basic requirement of scalability for search engines. In this paper, we present a real-time 3D shape search engine based on the projective images of 3D shapes. The real-time property of our search engine results from the following aspects: (1) efficient projection and view feature extraction using GPU acceleration; (2) the first inverted file, referred as F-IF, is utilized to speed up the procedure of multi-view matching; (3) the second inverted file (S-IF), which captures a local distribution of 3D shapes in the feature manifold, is adopted for efficient context-based reranking. As a result, for each query the retrieval task can be finished within one second despite the necessary cost of IO overhead. We name the proposed 3D shape search engine, which combines GPU acceleration and Inverted File (Twice), as GIFT. Besides its high efficiency, GIFT also outperforms the state-of-the-art methods significantly in retrieval accuracy on various shape benchmarks and competitions

We introduce G-CNN, an object detection technique based on CNNs which works without proposal algorithms. G-CNN starts with a multi-scale grid of fixed bounding boxes. We train a regressor to move and scale elements of the grid towards objects iteratively. G-CNN models the problem of object detection as finding a path from a fixed grid to boxes tightly surrounding the objects. G-CNN with around 180 boxes in a multi-scale grid performs comparably to Fast R-CNN which uses around 2K bounding boxes generated with a proposal technique. This strategy makes detection faster by removing the object proposal stage as well as reducing the number of boxes to be processed.

In this paper we present a method for computing dense correspondence between a set of 3D face meshes using functional maps. The functional maps paradigm brings with it a number of advantages for face correspondence. First, it allows us to combine various notions of correspondence. We do so by proposing a number of face-specific functions, suited to either within- or between-subject correspondence. Second, we propose a groupwise variant of the method allowing us to compute cycle-consistent functional maps between all faces in a training set. Since functional maps are of much lower dimension than point-to-point correspondences, this is feasible even when the input meshes are very high resolution. Finally, we show how a functional map provides a geometric constraint that can be used to filter feature matches between non-rigidly deforming surfaces.

We present a global optimization approach to optical flow estimation. The approach optimizes a classical optical flow objective over the full space of mappings between discrete grids. No descriptor matching is used. The highly regular structure of the space of mappings enables optimizations that reduce the computational complexity of the algorithm's inner loop from quadratic to linear and support efficient matching of tens of thousands of nodes to tens of thousands of displacements. We show that one-shot global optimization of a classical Horn-Schunck-type objective over regular grids at a single resolution is sufficient to initialize continuous interpolation and achieve state-of-the-art performance on challenging modern benchmarks.

Traditional image denoising algorithms always assume the noise to be homogeneous white Gaussian distributed. However, the noise on real images can be much more complex empirically. This paper addresses this problem and proposes a novel blind image denoising algorithm which can cope with real-world noisy images even when the noise model is not provided. It is realized by modeling image noise with mixture of Gaussian distribution (MoG) which can approximate large varieties of continuous distributions. As the number of components for MoG is unknown practically, this work adopts Bayesian nonparametric technique and proposes a novel Low-rank MoG filter (LR-MoG) to recover clean signals (patches) from noisy ones contaminated by MoG noise. Based on LR-MoG, a novel blind image denoising approach is developed. To test the proposed method, this study conducts extensive experiments on synthesis and real images. Our method achieves the state-of-the-art performance consistently.

We propose to summarize a video into a few key objects by selecting representative object proposals generated from video frames. This representative selection problem is formulated as a sparse dictionary selection problem, i.e., choosing a few representatives object proposals to reconstruct the whole proposal pool. Compared with existing sparse dictionary selection based representative selection methods, our new formulation can incorporate object proposal priors and locality prior in the feature space when selecting representatives. Consequently it can better locate key objects and suppress outlier proposals. We convert the optimization problem into a proximal gradient problem and solve it by the fast iterative shrinkage thresholding algorithm (FISTA). Experiments on synthetic data and real benchmark datasets show promising results of our key object summarization apporach in video content mining and search. Comparisons with existing representative selection approaches such as K-mediod, sparse dictionary selection and density based selection validate that our formulation can better capture the key video objects despite appearance variations, cluttered backgrounds and camera motions.

Internet photo collections naturally contain a large variety of illumination conditions, with the largest difference between day and night images. Current modeling techniques do not embrace the broad illumination range often leading to reconstruction failure or severe artifacts. We present an algorithm that leverages the appearance variety to obtain more complete and accurate scene geometry along with consistent multi-illumination appearance information. The proposed method relies on automatic scene appearance grouping, which is used to obtain separate dense 3D models. Subsequent model fusion combines the separate models into a complete and accurate reconstruction of the scene. In addition, we propose a method to derive the appearance information for the model under the different illumination conditions, even for scene parts that are not observed under one illumination condition. To achieve this, we develop a cross-illumination color transfer technique. We evaluate our method on a large variety of landmarks from across Europe reconstructed from a database of 7.4M images.

The rule of perspectivity that 'straight-lines-must-remain-straight' is easily inflected in CMOS cameras by distortions introduced by motion. Lines can be rendered as curves due to the row-wise exposure mechanism known as rolling shutter (RS). We solve the problem of correcting distortions arising from handheld cameras due to RS effect from a single image free from motion blur with special relevance to urban scenes. We develop a procedure to extract prominent curves from the RS image since this is essential for deciphering the varying row-wise motion. We pose an optimization problem with line desirability costs based on straightness, angle, and length, to resolve the geometric ambiguities while estimating the camera motion based on a rotation-only model assuming known camera intrinsic matrix. Finally, we rectify the RS image based on the estimated camera trajectory using inverse mapping. We show rectification results for RS images captured using mobile phone cameras. We also compare our single image method against existing video and nonblind RS rectification methods that typically require multiple images.

We investigate whether it is possible to improve the performance of automated facial forensic sketch matching by learning from examples of facial forgetting over time. Forensic facial sketch recognition is a key capability for law enforcement, but remains an unsolved problem. It is extremely challenging because there are three distinct contributors to the domain gap between forensic sketches and photos: The well studied sketch-photo modality gap, and the less studied gaps due to (i) the forgetting process of the eye-witness and (ii) their inability to elucidate their memory. In this paper we address the memory problem head on by introducing a database of 400 forensic sketches created at different time-delays. Based on this database we build a model to reverse the forgetting process. Surprisingly, we show that it is possible to systematically "un-forget" facial details. Moreover, it is possible to apply this model to dramatically improve forensic sketch recognition in practice: we achieve state of the art results when matching 195 benchmark forensic sketches against corresponding photos and a 10,030 mugshot database.

A first-person video can generate powerful physical sensations of action in an observer. In this paper, we focus on a problem of Force from Motion---decoding the sensation of 1) passive forces such as the gravity, 2) the physical scale of the motion (speed) and space, and 3) active forces exerted by the observer such as pedaling a bike or banking on a ski turn. The sensation of gravity can be observed in a natural image. We learn this image cue for predicting a gravity direction in a 2D image and integrate the prediction across images to estimate the 3D gravity direction using structure from motion. The sense of physical scale is revealed to us when the body is in a dynamically balanced state. We compute the unknown physical scale of 3D reconstructed camera motion by leveraging the torque equilibrium at a banked turn that relates the centripetal force, gravity, and the body leaning angle. The active force and torque governs 3D egomotion through the physics of rigid body dynamics. Using an inverse dynamics optimization, we directly minimize 2D reprojection error (in video) with respect to 3D world structure, active forces, and additional passive forces such as air drag and friction force. We use structure from motion with the physical scale and gravity direction as an initialization of our bundle adjustment for force estimation. Our method shows quantitatively equivalent reconstruction comparing to IMU measurements in terms of gravity and scale recovery and outperforms method based on 2D optical flow for an active action recognition task. We apply our method to first person videos of mountain biking, urban bike racing, skiing, speedflying with parachute, and wingsuit flying where inertial measurements are not accessible.

We present a fast, practical method for personalizing a hand shape basis to an individual user's detailed hand shape using only a small set of depth images. To achieve this, we minimize an energy based on a sum of render-and-compare cost functions called the golden energy. However, this energy is only piecewise continuous, due to pixels crossing occlusion boundaries, and is therefore not obviously amenable to efficient gradient-based optimization. A key insight is that the energy is the combination of a smooth low-frequency function with a high-frequency, low-amplitude, piecewise continuous function. A central finite difference approximation with a suitable step size can therefore jump over the discontinuities to obtain a good approximation to the energy's low-frequency behavior, allowing efficient gradient-based optimization. Experimental results quantitatively demonstrate for the first time that detailed personalized models improve the accuracy of hand tracking and achieve competitive results in both tracking and model registration.

We focus on the problem of wearer's action recognition in first person a.k.a. egocentric videos. This problem is more challenging than third person activity recognition due to unavailability of wearer's pose and sharp movements in the videos caused by the natural head motion of the wearer. Carefully crafted features based on hands and objects cues for the problem have been shown to be successful for limited targeted datasets. We propose convolutional neural networks (CNNs) for end to end learning and classification of wearer's actions. The proposed network makes use of egocentric cues by capturing hand pose, head motion and saliency map. It is compact. It can also be trained from relatively small number of labeled egocentric videos that are available. We show that the proposed network can generalize and give state of the art performance on various disparate egocentric action datasets.

Long training times for high-accuracy deep neural networks (DNNs) impede research into new DNN architectures and slow the development of high-accuracy DNNs. In this paper we present FireCaffe, which successfully scales deep neural network training across a cluster of GPUs. We also present a number of best practices to aid in comparing advancements in methods for scaling and accelerating the training of deep neural networks. The speed and scalability of distributed algorithms is almost always limited by the overhead of communicating between servers; DNN training is not an exception to this rule. Therefore, the key consideration here is to reduce communication overhead wherever possible, while not degrading the accuracy of the DNN models that we train. Our approach has three key pillars. First, we select network hardware that achieves high bandwidth between GPU servers -- Infiniband or Cray interconnects are ideal for this. Second, we consider a number of communication algorithms, and we find that reduction trees are more efficient and scalable than the traditional parameter server approach. Third, we optionally increase the batch size to reduce the total quantity of communication during DNN training, and we identify hyperparameters that allow us to reproduce the small-batch accuracy while training with large batch sizes. When training GoogLeNet and Network-in-Network on ImageNet, we achieve a 47x and 39x speedup, respectively, when training on a cluster of 128 GPUs.

Given a food image, can a fine-grained object recognition engine tell "which restaurant which dish" the food belongs to? Such ultra-fine grained image recognition is the key for many applications like search by images, but it is very challenging because it needs to discern subtle difference between classes while dealing with the scarcity of training data. Fortunately, the ultra-fine granularity naturally brings rich relationships among object classes. This paper proposes a novel approach to exploit the rich relationships through bipartite-graph labels (BGL). We show how to model BGL in an overall convolutional neural networks and the resulting system can be optimized through back-propagation. We also show that it is computationally efficient in inference thanks to the bipartite structure. To facilitate the study, we construct a new food benchmark dataset, which consists of 37,885 food images collected from 6 restaurants and totally 975 menus. Experimental results on this new food and three other datasets demonstrate BGL advances previous works in fine-grained object recognition. An online demo is available at http://www.f-zhou.com/fg_demo/.

Existing fine-grained visual categorization methods often suffer from three challenges: lack of training data, large number of fine-grained categories, and high intra-class vs. low inter-class variance. In this work we propose a generic iterative framework for fine-grained categorization and dataset bootstrapping that handles these three challenges. Using deep metric learning with humans in the loop, we learn a low dimensional feature embedding with anchor points on manifolds for each category. These anchor points capture intra-class variances and remain discriminative between classes. In each round, images with high confidence scores from our model are sent to humans for labeling. By comparing with exemplar images, labelers mark each candidate image as either a "true positive" or a "false positive." True positives are added into our current dataset and false positives are regarded as "hard negatives" for our metric learning model. Then the model is re-trained with an expanded dataset and hard negatives for the next round. To demonstrate the effectiveness of the proposed framework, we bootstrap a fine-grained flower dataset with 620 categories from Instagram images. The proposed deep metric learning scheme is evaluated on both our dataset and the CUB-200-2001 Birds dataset. Experimental evaluations show significant performance gain using dataset bootstrapping and demonstrate state-of-the-art results achieved by the proposed deep metric learning methods.

We present an approach to long-range spatio-temporal regularization in semantic video segmentation. Temporal regularization in video is challenging because both the camera and the scene may be in motion. Thus Euclidean distance in the space-time volume is not a good proxy for correspondence. We optimize the mapping of pixels to a Euclidean feature space so as to minimize distances between corresponding points. Structured prediction is performed by a dense CRF that operates on the optimized features. Experimental results demonstrate that the presented approach increases the accuracy and temporal consistency of semantic video segmentation.

The well-known word analogy experiments show that the recent word vectors capture fine-grained linguistic regularities in words by linear vector offsets, but it is unclear how well the simple vector offsets can encode visual regularities over words. We study a particular image-word relevance relation in this paper. Our results tell that, given an image, its relevant tags' word vectors rank ahead of the irrelevant tags' along a principal direction in the word vector space. Inspired by this observation, we propose to solve image tagging by estimating the principal direction for an image. Particularly, we exploit linear mappings and nonlinear deep neural networks to approximate the principal direction from an input image. We arrive at a quite versatile tagging model. It runs fast given a test image, in constant time w.r.t. the training set size. It not only gives rise to superior performance for the conventional tagging task on the NUS-WIDE dataset, but also outperforms competitive baselines on annotating images with previously unseen tags. To this end, we name our approach fast zero-shot image tagging (Fast0Tag) to recognize that it possesses the advantages of both FastTag (Chen et al. 2013) and zero-shot learning.

In this paper, we aim to learn a mapping (or embedding) from images to a compact binary space in which Hamming distances correspond to a ranking measure for the image retrieval task. We make use of a triplet loss because this has been shown to be most effective for ranking problems. How- ever, training in previous works can be prohibitively expensive due to the fact that optimization is directly performed on the triplet space, where the number of possible triplets for training is cubic in the number of training examples. To address this issue, we propose to formulate high-order binary codes learning as a multi-label classification problem by explicitly separating learning into two interleaved stages. To solve the first stage, we design a large-scale high-order binary codes inference algorithm to reduce the high-order objective to a standard binary quadratic problem such that graph cuts can be used to efficiently infer the binary codes which serve as the labels of each training datum. In the second stage we propose to map the original image to compact binary codes via carefully designed deep convolutional neural networks (CNNs) and the hash- ing function fitting can be solved by training binary CNN classifiers. An incremental/interleaved optimization strategy is proffered to ensure that these two steps are interactive with each other during training for better accuracy. We conduct experiments on several benchmark datasets, which demonstrate both improved training time (by as much as two orders of magnitude) as well as producing state-of-the- art hashing for various retrieval tasks.

In many large-scale video analysis scenarios, one is interested in localizing and recognizing human activities that occur in short temporal intervals within long untrimmed videos. Current approaches for activity detection still struggle to handle large-scale video collections and the task remains relatively unexplored. This is in part due to the computational complexity of current action recognition approaches and the lack of a method that proposes fewer intervals in the video, where activity processing can be focused. In this paper, we introduce a proposal method that aims to recover temporal segments containing actions in untrimmed videos. Building on techniques for learning sparse dictionaries, we introduce a learning framework to represent and retrieve activity proposals. We demonstrate the capabilities of our method in not only producing high quality proposals but also in its efficiency. Finally, we show the positive impact our method has on recognition performance when it is used for action detection, while running at 10FPS.