The ImageNet Large Scale Visual Recognition Challenge is a benchmark in object category classification and detection on hundreds of object categories and millions of images. The challenge has been run annually from 2010 to present, attracting participation from more than fifty institutions. This paper describes the creation of this benchmark dataset and the advances in object recognition that have been possible as a result. We discuss the challenges of collecting large-scale ground truth annotation, highlight key breakthroughs in categorical object recognition, provide a detailed analysis of the current state of the field of large-scale image classification and object detection, and compare the state-of-the-art computer vision accuracy with human accuracy. We conclude with lessons learned in the 5 years of the challenge, and propose future directions and improvements.

/pdf/imagenet-large-scale-visual-recognition-challenge-caky5nmmix.pdf

ImageNet Large Scale Visual Recognition Challenge

We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding. This is achieved by gathering images of complex everyday scenes containing common objects in their natural context. Objects are labeled using per-instance segmentations to aid in precise object localization. Our dataset contains photos of 91 objects types that would be easily recognizable by a 4 year old. With a total of 2.5 million labeled instances in 328k images, the creation of our dataset drew upon extensive crowd worker involvement via novel user interfaces for category detection, instance spotting and instance segmentation. We present a detailed statistical analysis of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide baseline performance analysis for bounding box and segmentation detection results using a Deformable Parts Model.

/pdf/microsoft-coco-common-objects-in-context-55ihxq1zt5.pdf

Microsoft COCO: Common Objects in Context

Convolutional Neural Networks define an exceptionally powerful class of models, but are still limited by the lack of ability to be spatially invariant to the input data in a computationally and parameter efficient manner. In this work we introduce a new learnable module, the Spatial Transformer, which explicitly allows the spatial manipulation of data within the network. This differentiable module can be inserted into existing convolutional architectures, giving neural networks the ability to actively spatially transform feature maps, conditional on the feature map itself, without any extra training supervision or modification to the optimisation process. We show that the use of spatial transformers results in models which learn invariance to translation, scale, rotation and more generic warping, resulting in state-of-the-art performance on several benchmarks, and for a number of classes of transformations.

/pdf/spatial-transformer-networks-459aor1dlh.pdf

Spatial transformer networks

In this work, we revisit the global average pooling layer proposed in [13], and shed light on how it explicitly enables the convolutional neural network (CNN) to have remarkable localization ability despite being trained on imagelevel labels. While this technique was previously proposed as a means for regularizing training, we find that it actually builds a generic localizable deep representation that exposes the implicit attention of CNNs on an image. Despite the apparent simplicity of global average pooling, we are able to achieve 37.1% top-5 error for object localization on ILSVRC 2014 without training on any bounding box annotation. We demonstrate in a variety of experiments that our network is able to localize the discriminative image regions despite just being trained for solving classification task1.

/pdf/learning-deep-features-for-discriminative-localization-2t8moexmea.pdf

Learning Deep Features for Discriminative Localization

We propose Prototypical Networks for the problem of few-shot classification, where a classifier must generalize to new classes not seen in the training set, given only a small number of examples of each new class. Prototypical Networks learn a metric space in which classification can be performed by computing distances to prototype representations of each class. Compared to recent approaches for few-shot learning, they reflect a simpler inductive bias that is beneficial in this limited-data regime, and achieve excellent results. We provide an analysis showing that some simple design decisions can yield substantial improvements over recent approaches involving complicated architectural choices and meta-learning. We further extend Prototypical Networks to zero-shot learning and achieve state-of-the-art results on the CU-Birds dataset.

/pdf/prototypical-networks-for-few-shot-learning-2l18ppn4sw.pdf

Prototypical Networks for Few-shot Learning

CUB-200-2011 is an extended version of CUB-200 [7], a challenging dataset of 200 bird species. The extended version roughly doubles the number of images per category and adds new part localization annotations. All images are annotated with bounding boxes, part locations, and at- tribute labels. Images and annotations were filtered by mul- tiple users of Mechanical Turk. We introduce benchmarks and baseline experiments for multi-class categorization and part localization.

/pdf/the-caltech-ucsd-birds-200-2011-dataset-430natao92.pdf

The Caltech-UCSD Birds-200-2011 Dataset

Caltech-UCSD Birds 200 (CUB-200) is a challenging image dataset annotated with 200 bird species. It was created to enable the study of subordinate categorization, which is not possible with other popular datasets that focus on basic level categories (such as PASCAL VOC, Caltech-101, etc). The images were downloaded from the website Flickr and filtered by workers on Amazon Mechanical Turk. Each image is annotated with a bounding box, a rough bird segmentation, and a set of attribute labels.

Caltech-UCSD Birds 200

Distributing labeling tasks among hundreds or thousands of annotators is an increasingly important method for annotating large datasets. We present a method for estimating the underlying value (e.g. the class) of each image from (noisy) annotations provided by multiple annotators. Our method is based on a model of the image formation and annotation process. Each image has different characteristics that are represented in an abstract Euclidean space. Each annotator is modeled as a multidimensional entity with variables representing competence, expertise and bias. This allows the model to discover and represent groups of annotators that have different sets of skills and knowledge, as well as groups of images that differ qualitatively. We find that our model predicts ground truth labels on both synthetic and real data more accurately than state of the art methods. Experiments also show that our model, starting from a set of binary labels, may discover rich information, such as different "schools of thought" amongst the annotators, and can group together images belonging to separate categories.

/pdf/the-multidimensional-wisdom-of-crowds-35s8gd1f1u.pdf

The Multidimensional Wisdom of Crowds

We present an interactive, hybrid human-computer method for object classification. The method applies to classes of objects that are recognizable by people with appropriate expertise (e.g., animal species or airplane model), but not (in general) by people without such expertise. It can be seen as a visual version of the 20 questions game, where questions based on simple visual attributes are posed interactively. The goal is to identify the true class while minimizing the number of questions asked, using the visual content of the image. We introduce a general framework for incorporating almost any off-the-shelf multi-class object recognition algorithm into the visual 20 questions game, and provide methodologies to account for imperfect user responses and unreliable computer vision algorithms. We evaluate our methods on Birds-200, a difficult dataset of 200 tightly-related bird species, and on the Animals With Attributes dataset. Our results demonstrate that incorporating user input drives up recognition accuracy to levels that are good enough for practical applications, while at the same time, computer vision reduces the amount of human interaction required.

/pdf/visual-recognition-with-humans-in-the-loop-2myqc00150.pdf

Visual recognition with humans in the loop

We propose an architecture for fine-grained visual categorization that approaches expert human performance in the classification of bird species. Our architecture first computes an estimate of the object's pose; this is used to compute local image features which are, in turn, used for classification. The features are computed by applying deep convolutional nets to image patches that are located and normalized by the pose. We perform an empirical study of a number of pose normalization schemes, including an investigation of higher order geometric warping functions. We propose a novel graph-based clustering algorithm for learning a compact pose normalization space. We perform a detailed investigation of state-of-the-art deep convolutional feature implementations and fine-tuning feature learning for fine-grained classification. We observe that a model that integrates lower-level feature layers with pose-normalized extraction routines and higher-level feature layers with unaligned image features works best. Our experiments advance state-of-the-art performance on bird species recognition, with a large improvement of correct classification rates over previous methods (75% vs. 55-65%).

/pdf/bird-species-categorization-using-pose-normalized-deep-2wmw0y5y96.pdf

Steve Branson

Papers

The Caltech-UCSD Birds-200-2011 Dataset

Caltech-UCSD Birds 200

The Multidimensional Wisdom of Crowds

Visual recognition with humans in the loop

Bird Species Categorization Using Pose Normalized Deep Convolutional Nets