TL;DR: A model that decomposes both images and paragraphs into their constituent parts is developed, detecting semantic regions in images and using a hierarchical recurrent neural network to reason about language.
Abstract: Recent progress on image captioning has made it possible to generate novel sentences describing images in natural language, but compressing an image into a single sentence can describe visual content in only coarse detail. While one new captioning approach, dense captioning, can potentially describe images in finer levels of detail by captioning many regions within an image, it in turn is unable to produce a coherent story for an image. In this paper we overcome these limitations by generating entire paragraphs for describing images, which can tell detailed, unified stories. We develop a model that decomposes both images and paragraphs into their constituent parts, detecting semantic regions in images and using a hierarchical recurrent neural network to reason about language. Linguistic analysis confirms the complexity of the paragraph generation task, and thorough experiments on a new dataset of image and paragraph pairs demonstrate the effectiveness of our approach.
TL;DR: In this article, the authors introduce a new model that is able to identify all events in a single pass of the video while simultaneously describing the detected events with natural language, which is called ActivityNet Captions.
Abstract: Most natural videos contain numerous events. For example, in a video of a “man playing a piano”, the video might also contain “another man dancing” or “a crowd clapping”. We introduce the task of dense-captioning events, which involves both detecting and describing events in a video. We propose a new model that is able to identify all events in a single pass of the video while simultaneously describing the detected events with natural language. Our model introduces a variant of an existing proposal module that is designed to capture both short as well as long events that span minutes. To capture the dependencies between the events in a video, our model introduces a new captioning module that uses contextual information from past and future events to jointly describe all events. We also introduce ActivityNet Captions, a large-scale benchmark for dense-captioning events. ActivityNet Captions contains 20k videos amounting to 849 video hours with 100k total descriptions, each with its unique start and end time. Finally, we report performances of our model for dense-captioning events, video retrieval and localization.
TL;DR: This article proposed a new framework based on conditional generative adversarial networks (CGAN), which jointly learns a generator to produce descriptions conditioned on images and an evaluator to assess how well a description fits the visual content.
Abstract: Despite the substantial progress in recent years, the image captioning techniques are still far from being perfect. Sentences produced by existing methods, e.g. those based on RNNs, are often overly rigid and lacking in variability. This issue is related to a learning principle widely used in practice, that is, to maximize the likelihood of training samples. This principle encourages high resemblance to the “ground-truth” captions, while suppressing other reasonable descriptions. Conventional evaluation metrics, e.g. BLEU and METEOR, also favor such restrictive methods. In this paper, we explore an alternative approach, with the aim to improve the naturalness and diversity – two essential properties of human expression. Specifically, we propose a new framework based on Conditional Generative Adversarial Networks (CGAN), which jointly learns a generator to produce descriptions conditioned on images and an evaluator to assess how well a description fits the visual content. It is noteworthy that training a sequence generator is nontrivial. We overcome the difficulty by Policy Gradient, a strategy stemming from Reinforcement Learning, which allows the generator to receive early feedback along the way. We tested our method on two large datasets, where it performed competitively against real people in our user study and outperformed other methods on various tasks.
TL;DR: In this article, the authors propose a new model that is able to identify all events in a single pass of the video while simultaneously describing the detected events with natural language, which can capture both short and long events that span minutes.
Abstract: Most natural videos contain numerous events. For example, in a video of a "man playing a piano", the video might also contain "another man dancing" or "a crowd clapping". We introduce the task of dense-captioning events, which involves both detecting and describing events in a video. We propose a new model that is able to identify all events in a single pass of the video while simultaneously describing the detected events with natural language. Our model introduces a variant of an existing proposal module that is designed to capture both short as well as long events that span minutes. To capture the dependencies between the events in a video, our model introduces a new captioning module that uses contextual information from past and future events to jointly describe all events. We also introduce ActivityNet Captions, a large-scale benchmark for dense-captioning events. ActivityNet Captions contains 20k videos amounting to 849 video hours with 100k total descriptions, each with it's unique start and end time. Finally, we report performances of our model for dense-captioning events, video retrieval and localization.
TL;DR: A comprehensive review of state-of-the-art deep learning approaches that have been used in the context of histopathological image analysis can be found in this paper, where a survey of over 130 papers is presented.
TL;DR: The authors proposed a unified learning framework that collectively addresses all the above issues by composing a committee of discriminators that can guide a base RNN generator towards more globally coherent generations, and human evaluation demonstrates that text generated by their model is preferred over that of baselines by a large margin, significantly enhancing the overall coherence, style, and information of the generations.
Abstract: Despite their local fluency, long-form text generated from RNNs is often generic, repetitive, and even self-contradictory. We propose a unified learning framework that collectively addresses all the above issues by composing a committee of discriminators that can guide a base RNN generator towards more globally coherent generations. More concretely, discriminators each specialize in a different principle of communication, such as Grice’s maxims, and are collectively combined with the base RNN generator through a composite decoding objective. Human evaluation demonstrates that text generated by our model is preferred over that of baselines by a large margin, significantly enhancing the overall coherence, style, and information of the generations.
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Abstract: We introduce Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments. The method is straightforward to implement, is computationally efficient, has little memory requirements, is invariant to diagonal rescaling of the gradients, and is well suited for problems that are large in terms of data and/or parameters. The method is also appropriate for non-stationary objectives and problems with very noisy and/or sparse gradients. The hyper-parameters have intuitive interpretations and typically require little tuning. Some connections to related algorithms, on which Adam was inspired, are discussed. We also analyze the theoretical convergence properties of the algorithm and provide a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework. Empirical results demonstrate that Adam works well in practice and compares favorably to other stochastic optimization methods. Finally, we discuss AdaMax, a variant of Adam based on the infinity norm.
111,197 citations
"A Hierarchical Approach for Generat..." refers methods in this paper
...Training is done via stochastic gradient descent with Adam [13] learning rate updates, implemented in Torch....
[...]
...Training is done via stochastic gradient descent with Adam [14], implemented in Torch....
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Abstract: Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O. 1. Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.
72,897 citations
"A Hierarchical Approach for Generat..." refers background or methods in this paper
...All baseline neural language models use two layers of LSTM [9] units with 512 dimensions....
[...]
...Alternative recurrent architectures such as long-short term memory (LSTM) [8] help alleviate this problem through a gating mechanism that improves gradient flow....
[...]
...At each timestep the hidden state of the last LSTM layer is used to predict a distribution over the words in the vocabulary, and a special END token signals the end of a sentence....
[...]
...We adopt the standard LSTM architecture [9] for both the word RNN and sentence RNN....
[...]
...Second, the hidden state hi is fed through a two-layer fullyconnected network to produce the topic vector ti ∈ RP for the ith sentence of the paragraph, which is the direct input to the word RNN.
Word RNN The word RNN is a two-layer LSTM with hidden size H = 512, which, given a topic vector ti ∈ RP from the sentence RNN, is responsible for generating the words of a sentence....
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.
TL;DR: 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 by gathering images of complex everyday scenes containing common objects in their natural context.
Abstract: 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.
30,462 citations
"A Hierarchical Approach for Generat..." refers background or methods in this paper
...Our hierarchical method also had a much wider vocabulary compared to the Template approach, though Sentence-Concat, trained on hundreds of thousands of MS COCO [20] captions, is a bit larger....
[...]
...Sentence-Concat: To demonstrate the difference between sentence-level and paragraph captions, this baseline samples and concatenates five sentence captions from a model [12] trained on MS COCO captions [20]....
[...]
...With the advent of large datasets pairing images with natural language descriptions [20, 34, 10, 16] it has recently become possible to generate novel sentences describing images [4, 6, 12, 22, 30]....
[...]
...To validate our method, we collected a dataset of image and paragraph pairs, which complements the whole-image and region-level annotations of MS COCO [20] and Visual Genome [16]....
[...]
...To what extent does describing images with paragraphs differ from sentence-level captioning? To answer this question, we collected a novel dataset of paragraph annotations, comparised of 19,551 MS COCO [20] and Visual Genome [16] images, where each image has been annotated with a paragraph description....
TL;DR: This paper proposed a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run.
Abstract: Human evaluations of machine translation are extensive but expensive. Human evaluations can take months to finish and involve human labor that can not be reused. We propose a method of automatic machine translation evaluation that is quick, inexpensive, and language-independent, that correlates highly with human evaluation, and that has little marginal cost per run. We present this method as an automated understudy to skilled human judges which substitutes for them when there is need for quick or frequent evaluations.