ICCV 2015 Papers

A common issue in deformable object detection is finding a good way to position the parts. This issue is even more outspoken when considering detection and pose estimation for 3D objects, where parts should be placed in a three-dimensional space. Some methods extract the 3D shape of the object from 3D CAD models. This limits their applicability to categories for which such models are available. Others represent the object with a predefined and simple shape (e.g. a cuboid). This extends the applicability of the model, but in many cases the pre-defined shape is too simple to properly represent the object in 3D. In this paper we propose a new method for the detection and pose estimation of 3D objects, that does not use any 3D CAD model or other 3D information. Starting from a simple and general 3D shape, we learn in a weakly supervised manner the 3D part locations that best fit the training data. As this method builds on a iterative estimation of the part locations, we introduce several speedups to make the method fast enough for practical experiments. We evaluate our model for the detection and pose estimation of faces and cars. Our method obtains results comparable with the state of the art, it is faster than most of the other approaches and does not need any additional 3D information.

With the rapid proliferation of smart mobile devices, users now take millions of photos every day. These include large numbers of clothing and accessory images. We would like to answer questions like `What outfit goes well with this pair of shoes?' To answer these types of questions, one has to go beyond learning visual similarity and learn a visual notion of compatibility across categories. In this paper, we propose a novel learning framework to help answer these types of questions. The main idea of this framework is to learn a feature transformation from images of items into a latent space that expresses compatibility. For the feature transformation, we use a Siamese Convolutional Neural Network (CNN) architecture, where training examples are pairs of items that are either compatible or incompatible. We model compatibility based on co-occurrence in large-scale user behavior data; in particular co-purchase data from Amazon.com. To learn cross-category fit, we introduce a strategic method to sample training data, where pairs of items are heterogeneous dyads, i.e., the two elements of a pair belong to different high-level categories. While this approach is applicable to a wide variety of settings, we focus on the representative problem of learning compatible clothing style. Our results indicate that the proposed framework is capable of learning semantic information about visual style and is able to generate outfits of clothes, with items from different categories, that go well together.

In this work, we develop a novel method for automatically learning aspects of the structure of a deep model, in order to improve its performance, especially when labeled training data are scarce. We propose a new convolutional neural network model with the Indian Buffet Process (IBP) prior, termed ibpCNN. The ibpCNN automatically adapts its structure to provided training data, achieves an optimal balance among model complexity, data fidelity and training loss, and thus offers better generalization performance. The proposed ibpCNN captures complicated data distribution in an unsupervised generative way. Therefore, ibpCNN can exploit unlabeled data -- which can be collected at low cost -- to learn its structure. After determining the structure, ibpCNN further learns its parameters according to specified tasks, in an end-to-end fashion, and produces discriminative yet compact representations. We evaluate the performance of ibpCNN, on fully- and semi-supervised image classification tasks; ibpCNN surpasses standard CNN models on benchmark datasets, with much smaller size and higher efficiency.

In this paper, we propose to learn temporal embeddings of video frames for complex video analysis. Large quantities of unlabeled video data can be easily obtained from the Internet. These videos possess the implicit weak label that they are sequences of temporally and semantically coherent images. We leverage this information to learn temporal embeddings for video frames by associating frames with the temporal context that they appear in. To do this, we propose a scheme for incorporating temporal context based on past and future frames in videos, and compare this to other contextual representations. In addition, we show how data augmentation using multi-resolution samples and hard negatives helps to significantly improve the quality of the learned embeddings. We evaluate various design decisions for learning temporal embeddings, and show that our embeddings can improve performance for multiple video tasks such as retrieval, classification, and temporal order recovery in unconstrained Internet video.

We propose a simple, yet effective approach for spatiotemporal feature learning using deep 3-dimensional convolutional networks (3D ConvNets) trained on a large scale supervised video dataset. Our findings are three-fold: 1) 3D ConvNets are more suitable for spatiotemporal feature learning compared to 2D ConvNets; 2) A homogeneous architecture with small 3x3x3 convolution kernels in all layers is among the best performing architectures for 3D ConvNets; and 3) Our learned features, namely C3D (Convolutional 3D), with a simple linear classifier outperform state-of-the-art methods on 4 different benchmarks and are comparable with current best methods on the other 2 benchmarks. In addition, the features are compact: achieving 52.8% accuracy on UCF101 dataset with only 10 dimensions and also very efficient to compute due to the fast inference of ConvNets. Finally, they are conceptually very simple and easy to train and use.

Robust and accurate visual tracking is one of the most challenging computer vision problems. Due to the inherent lack of training data, a robust approach for constructing a target appearance model is crucial. Recently, discriminatively learned correlation filters (DCF) have been successfully applied to address this problem for tracking. These methods utilize a periodic assumption of the training samples to efficiently learn a classifier on all patches in the target neighborhood. However, the periodic assumption also introduces unwanted boundary effects, which severely degrade the quality of the tracking model. We propose Spatially Regularized Discriminative Correlation Filters (SRDCF) for tracking. A spatial regularization component is introduced in the learning to penalize correlation filter coefficients depending on their spatial location. Our SRDCF formulation allows the correlation filters to be learned on a significantly larger set of negative training samples, without corrupting the positive samples. We further propose an optimization strategy, based on the iterative Gauss-Seidel method, for efficient online learning of our SRDCF. Experiments are performed on four benchmark datasets: OTB-2013, ALOV++, OTB-2015, and VOT2014. Our approach achieves state-of-the-art results on all four datasets. On OTB-2013 and OTB-2015, we obtain an absolute gain of 8.0% and 8.2% respectively, in mean overlap precision, compared to the best existing trackers.

Social relation defines the association, e.g., warm, friendliness, and dominance, between two or more people. Motivated by psychological studies, we investigate if such fine grained and high-level relation traits can be characterised and quantified from face images in the wild. To address this challenging problem we propose a deep model that learns a rich face representation to capture gender, expression, head pose, and age-related attributes, and then performs pairwise-face reasoning for relation prediction. To learn from heterogeneous attribute sources, we formulate a new network architecture with a bridging layer to leverage the inherent correspondences among these datasets. It can also cope with missing target attribute labels. Extensive experiments show that our approach is effective for fine-grained social relation learning in images and videos.

In this paper, we address the problem of simultaneously recovering the 3D shape and pose of a deformable and potentially elastic object from 2D motion. This is a highly ambiguous problem typically tackled by using low-rank shape and trajectory constraints. We show that formulating the problem in terms of a low-rank force space that induces the deformation, allows for a better physical interpretation of the resulting priors and a more accurate representation of the actual object's behavior. However, this comes at the price of, besides force and pose, having to estimate the elastic model of the object. For this, we use an Expectation Maximization strategy, where each of these parameters are successively learned within partial M-steps, while robustly dealing with missing observations. We thoroughly validate the approach on both mocap and real sequences, showing more accurate 3D reconstructions than state-of-the-art, and additionally providing an estimate of the full elastic model with no a priori information.

State of the art approaches for Semi-Supervised Learning (SSL) usually follow a two-stage framework -- constructing an affinity matrix from the data and then propagating the partial labels on this affinity matrix to infer those unknown labels. While such a two-stage framework has been successful in many applications, solving two subproblems separately only once is still suboptimal because it does not fully exploit the correlation between the affinity and the labels. In this paper, we formulate the two stages of SSL into a unified optimization framework, which learns both the affinity matrix and the unknown labels simultaneously. In the unified framework, both the given labels and the estimated labels are used to learn the affinity matrix and to infer the unknown labels. We solve the unified optimization problem via an alternating direction method of multipliers combined with label propagation. Extensive experiments on a synthetic data set and several benchmark data sets demonstrate the effectiveness of our approach.

Existing approaches toward Image super-resolution (SR) is often either data-driven (e.g., based on internet-scale matching and web image retrieval) or model-based (e.g., formulated as an Maximizing a Posterior estimation problem). The former is conceptually simple yet heuristic; while the latter is constrained by the fundamental limit of frequency aliasing. In this paper, we propose to develop a hybrid approach toward SR by combining those two lines of ideas. More specifically, the parameters underlying sparse distributions of desirable HR image patches are learned from a pair of LR image and retrieved HR images. Our hybrid approach can be interpreted as the first attempt of reconciling the difference between parametric and nonparametric models for low-level vision tasks. Experimental results show that the proposed hybrid SR method performs much better than existing state-of-the-art methods in terms of both subjective and objective image qualities.

We propose a framework that infers mid-level visual properties of an image by learning about ordinal relation- ships. Instead of estimating metric quantities directly, the system proposes pairwise relationship estimates for points in the input image. These sparse probabilistic ordinal mea- surements are globalized to create a dense output map of continuous metric measurements. Estimating order rela- tionships between pairs of points has several advantages over metric estimation: it solves a simpler problem than metric regression; humans are better at relative judgements, so data collection is easier; ordinal relationships are invari- ant to monotonic transformations of the data, thereby in- creasing the robustness of the system and providing qualitatively different information. We demonstrate that this frame- work works well on two important mid-level vision tasks: intrinsic image decomposition and depth from an RGB im- age. We train two systems with the same architecture on data from these two modalities. We provide an analysis of the resulting models, showing that they learn a number of simple rules to make ordinal decisions. We apply our algo-rithm to depth estimation, with good results, and intrinsic image decomposition, with state-of-the-art results.

This paper presents the idea of learning optimal filters for color image reconstruction based on a novel concept of nonlinear spectral image decompositions recently proposed by Guy Gilboa. We use a multiscale image decomposition approach based on total variation regularization and Bregman iterations to represent the input data as the sum of image layers containing features at different scales. Filtered images can be obtained by weighted linear combinations of the different frequency layers. We introduce the idea of learning optimal filters for the task of image denoising, and propose the idea of mixing high frequency components of different color channels. Our numerical experiments demonstrate that learning the optimal weights can significantly improve the results in comparison to the standard variational approach, and achieves state-of-the-art image denoising results.

In this paper, we address the task of learning novel visual concepts, and their interactions with other concepts, from a few images with sentence descriptions. Using linguistic context and visual features, our method is able to efficiently hypothesize the semantic meaning of new words and add them to its word dictionary so that they can be used to describe images which contain these novel concepts. Our method has an image captioning module based on the m-RNN model with several improvements. In particular, we propose a transposed weight sharing scheme, which not only improves performance on image captioning, but also makes the model more suitable for the novel concept learning task. We propose methods to prevent overfitting the new concepts. In addition, three novel concept datasets are constructed for this new task, and are publicly available on the project page. In the experiments, we show that our method effectively learns novel visual concepts from a few examples without disturbing the previously learned concepts. The project page is: http://www.stat.ucla.edu/ junhua.mao/projects/child_learning.html

We describe an automated method for image colorization that learns to colorize from examples. Our method exploits a LEARCH framework to train a quadratic objective function in the chromaticity maps, comparable to a Gaussian random field. The coefficients of the objective function are conditioned on image features, using a random forest. The objective function admits correlations on long spatial scales, and can control spatial error in the colorization of the image. Images are then colorized by minimizing this objective function. We demonstrate that our method strongly outperforms a natural baseline on large-scale experiments with images of real scenes using a demanding loss function. We demonstrate that learning a model that is conditioned on scene produces improved results. We show how to incorporate a desired color histogram into the objective function, and that doing so can lead to further improvements in results.

In this paper, we introduce new edge-based features for the task of recovering the 3D layout of an indoor scene from a single image. Indoor scenes have certain edges that are very informative about the spatial layout of the room, namely, the edges formed by the pairwise intersections of room faces (two walls, wall and ceiling, wall and floor). In contrast with previous approaches that rely on area-based features like geometric context and orientation maps, our method attempts to directly detect these informative edges. We learn to predict 'informative edge' probability maps using two recent methods that exploit local and global context, respectively: structured edge detection forests, and a fully convolutional network for pixelwise labeling. We show that the fully convolutional network is quite successful at predicting the informative edges even when they lack contrast or are occluded, and that the accuracy can be further improved by training the network to jointly predict the edges and the geometric context. Using features derived from the 'informative edge' maps, we learn a maximum margin structured classifier that achieves state-of-the-art performance on layout prediction.

Good representations of data do help in many machine learning tasks such as recommendation. It is often a great challenge for traditional recommender systems to learn representative features of both users and images in large social networks, in particular, social curation networks, which are characterized as the extremely sparse links between users and images, and the extremely diverse visual contents of images. To address the challenges, we propose a novel deep model which learns the unified feature representations for both users and images. This is done by transforming the heterogeneous user-image networks into homogeneous low-dimensional representations, which facilitate a recommender to trivially recommend images to users by feature similarity. We also develop a fast online algorithm that can be easily scaled up to large networks in an asynchronously parallel way. We conduct extensive experiments on a representative subset of Pinterest, containing 1,456,540 images and 1,000,000 users. Results of image recommendation experiments demonstrate that our feature learning approach significantly outperforms other state-of-the-art recommendation methods.

Many visual recognition tasks involve modeling variables which are structurally related. Hidden conditional random fields (HCRFs) are a powerful class of models for encoding structure in weakly supervised training examples. This paper presents HCRF-Boost, a novel and general framework for learning HCRFs in functional space. An algorithm is proposed to learn the potential functions of an HCRF as a combination of abstract nonlinear feature functions, expressed by regression models. Consequently, the resulting latent structured model is not restricted to traditional log-linear potential functions or any explicit parameterization. Further, functional optimization helps to avoid direct interactions with the possibly large parameter space of nonlinear models and improves efficiency. As a result, a complex and flexible ensemble method is achieved for structured prediction which can be successfully used in a variety of applications. We validate the effectiveness of this method on tasks such as group activity recognition, human action recognition, and multi-instance learning of video events.

Learning strong feature representations from large scale supervision has achieved remarkable success in computer vision as the emergence of deep learning techniques. It is driven by big visual data with rich annotations. This paper contributes a large-scale object attribute database (The dataset is available on www.ee.cuhk.edu.hk/ xgwang/ImageNetAttribute.html) that contains rich attribute annotations (over 300 attributes) for ~180k samples and 494 object classes. Based on the ImageNet object detection dataset, it annotates the rotation, viewpoint, object part location, part occlusion, part existence, common attributes, and class-specific attributes. Then we use this dataset to train deep representations and extensively evaluate how these attributes are useful on the general object detection task. In order to make better use of the attribute annotations, a deep learning scheme is proposed by modeling the relationship of attributes and hierarchically clustering them into semantically meaningful mixture types. Experimental results show that the attributes are helpful in learning better features and improving the object detection accuracy by 2.6% in mAP on the ILSVRC 2014 object detection dataset and 2.4% in mAP on PASCAL VOC 2007 object detection dataset. Such improvement is well generalized across datasets.

Crowdsourced 3D CAD models are easily accessible online, and can potentially generate an infinite number of training images for almost any object category. We show that augmenting the training data of contemporary Deep Convolutional Neural Net (DCNN) models with such synthetic data can be effective, especially when real training data is limited or not well matched to the target domain. Most freely available CAD models capture 3D shape but are often missing other low level cues, such as realistic object texture, pose, or background. In a detailed analysis, we use synthetic CAD images to probe the ability of DCNN to learn without these cues, with surprising findings. In particular, we show that when the DCNN is fine-tuned on the target detection task, it exhibits a large degree of invariance to missing low-level cues, but, when pretrained on generic ImageNet classification, it learns better when the low-level cues are simulated. We show that our synthetic DCNN training approach significantly outperforms previous methods on the benchmark PASCAL VOC2007 dataset when learning in the few-shot scenario and improves performance in a domain shift scenario on the Office benchmark.