The success of deep learning in vision can be attributed to: (a) models with high capacity; (b) increased computational power; and (c) availability of large-scale labeled data. Since 2012, there have been significant advances in representation capabilities of the models and computational capabilities of GPUs. But the size of the biggest dataset has surprisingly remained constant. What will happen if we increase the dataset size by 10× or 100×? This paper takes a step towards clearing the clouds of mystery surrounding the relationship between 'enormous data' and visual deep learning. By exploiting the JFT-300M dataset which has more than 375M noisy labels for 300M images, we investigate how the performance of current vision tasks would change if this data was used for representation learning. Our paper delivers some surprising (and some expected) findings. First, we find that the performance on vision tasks increases logarithmically based on volume of training data size. Second, we show that representation learning (or pretraining) still holds a lot of promise. One can improve performance on many vision tasks by just training a better base model. Finally, as expected, we present new state-of-theart results for different vision tasks including image classification, object detection, semantic segmentation and human pose estimation. Our sincere hope is that this inspires vision community to not undervalue the data and develop collective efforts in building larger datasets.

We address the problem of distance metric learning (DML), defined as learning a distance consistent with a notion of semantic similarity. Traditionally, for this problem supervision is expressed in the form of sets of points that follow an ordinal relationship -an anchor point x is similar to a set of positive points Y , and dissimilar to a set of negative points Z, and a loss defined over these distances is minimized. While the specifics of the optimization differ, in this work we collectively call this type of supervision Triplets and all methods that follow this pattern Triplet-Based methods. These methods are challenging to optimize. A main issue is the need for finding informative triplets, which is usually achieved by a variety of tricks such as increasing the batch size, hard or semi-hard triplet mining, etc. Even with these tricks, the convergence rate of such methods is slow. In this paper we propose to optimize the triplet loss on a different space of triplets, consisting of an anchor data point and similar and dissimilar proxy points which are learned as well. These proxies approximate the original data points, so that a triplet loss over the proxies is a tight upper bound of the original loss. This proxy-based loss is empirically better behaved. As a result, the proxy-loss improves on state-of-art results for three standard zero-shot learning datasets, by up to 15% points, while converging three times as fast as other triplet-based losses.

The goal of this paper is to discover a set of discriminative patches which can serve as a fully unsupervised mid-level visual representation. The desired patches need to satisfy two requirements: 1) to be representative, they need to occur frequently enough in the visual world; 2) to be discriminative, they need to be different enough from the rest of the visual world. The patches could correspond to parts, objects, "visual phrases", etc. but are not restricted to be any one of them. We pose this as an unsupervised discriminative clustering problem on a huge dataset of image patches. We use an iterative procedure which alternates between clustering and training discriminative classifiers, while applying careful cross-validation at each step to prevent overfitting. The paper experimentally demonstrates the effectiveness of discriminative patches as an unsupervised mid-level visual representation, suggesting that it could be used in place of visual words for many tasks. Furthermore, discriminative patches can also be used in a supervised regime, such as scene classification, where they demonstrate state-of-the-art performance on the MIT Indoor-67 dataset.

Figure 1: These two photos might seem nondescript, but each contains hints about which city it might belong to. Given a large image database of a given city, our algorithm is able to automatically discover the geographically-informative elements (patch clusters to the right of each photo) that help in capturing its "look and feel". On the left, the emblematic street sign, a balustrade window, and the balcony support are all very indicative of Paris, while on the right, the neoclassical columned entryway sporting a balcony, a Victorian window, and, of course, the cast iron railing are very much features of London. AbstractGiven a large repository of geotagged imagery, we seek to automatically find visual elements, e.g. windows, balconies, and street signs, that are most distinctive for a certain geo-spatial area, for example the city of Paris. This is a tremendously difficult task as the visual features distinguishing architectural elements of different places can be very subtle. In addition, we face a hard search problem: given all possible patches in all images, which of them are both frequently occurring and geographically informative? To address these issues, we propose to use a discriminative clustering approach able to take into account the weak geographic supervision. We show that geographically representative image elements can be discovered automatically from Google Street View imagery in a discriminative manner. We demonstrate that these elements are visually interpretable and perceptually geo-informative. The discovered visual elements can also support a variety of computational geography tasks, such as mapping architectural correspondences and influences within and across cities, finding representative elements at different geo-spatial scales, and geographically-informed image retrieval.

We present a method that learns to answer visual questions by selecting image regions relevant to the text-based query. Our method maps textual queries and visual features from various regions into a shared space where they are compared for relevance with an inner product. Our method exhibits significant improvements in answering questions such as "what color," where it is necessary to evaluate a specific location, and "what room," where it selectively identifies informative image regions. Our model is tested on the recently released VQA [1] dataset, which features free-form human-annotated questions and answers.

Abstract. We consider the problem of semi-supervised bootstrap learning for scene categorization. Existing semi-supervised approaches are typically unreliable and face semantic drift because the learning task is under-constrained. This is primarily because they ignore the strong interactions that often exist between scene categories, such as the common attributes shared across categories as well as the attributes which make one scene different from another. The goal of this paper is to exploit these relationships and constrain the semi-supervised learning problem. For example, the knowledge that an image is an auditorium can improve labeling of amphitheaters by enforcing constraint that an amphitheater image should have more circular structures than an auditorium image. We propose constraints based on mutual exclusion, binary attributes and comparative attributes and show that they help us to constrain the learning problem and avoid semantic drift. We demonstrate the effectiveness of our approach through extensive experiments, including results on a very large dataset of one million images.

Given a large repository of geo-tagged imagery, we seek to automatically find visual elements, for example windows, balconies, and street signs, that are most distinctive for a certain geo-spatial area, for example the city of Paris. This is a tremendously difficult task as the visual features distinguishing architectural elements of different places can be very subtle. In addition, we face a hard search problem: given all possible patches in all images, which of them are both frequently occurring and geographically informative? To address these issues, we propose to use a discriminative clustering approach able to take into account the weak geographic supervision. We show that geographically representative image elements can be discovered automatically from Google Street View imagery in a discriminative manner. We demonstrate that these elements are visually interpretable and perceptually geo-informative. The discovered visual elements can also support a variety of computational geography tasks, such as mapping architectural correspondences and influences within and across cities, finding representative elements at different geo-spatial scales, and geographically informed image retrieval.

No abstract