YouTube2Text: Recognizing and Describing Arbitrary Activities Using Semantic Hierarchies and Zero-Shot Recognition
01 Dec 2013-pp 2712-2719
TL;DR: This paper presents a solution that takes a short video clip and outputs a brief sentence that sums up the main activity in the video, such as the actor, the action and its object, and uses a Web-scale language model to ``fill in'' novel verbs.
Abstract: Despite a recent push towards large-scale object recognition, activity recognition remains limited to narrow domains and small vocabularies of actions. In this paper, we tackle the challenge of recognizing and describing activities ``in-the-wild''. We present a solution that takes a short video clip and outputs a brief sentence that sums up the main activity in the video, such as the actor, the action and its object. Unlike previous work, our approach works on out-of-domain actions: it does not require training videos of the exact activity. If it cannot find an accurate prediction for a pre-trained model, it finds a less specific answer that is also plausible from a pragmatic standpoint. We use semantic hierarchies learned from the data to help to choose an appropriate level of generalization, and priors learned from Web-scale natural language corpora to penalize unlikely combinations of actors/actions/objects, we also use a Web-scale language model to ``fill in'' novel verbs, i.e. when the verb does not appear in the training set. We evaluate our method on a large YouTube corpus and demonstrate it is able to generate short sentence descriptions of video clips better than baseline approaches.
Citations
More filters
07 Jun 2015
TL;DR: A novel recurrent convolutional architecture suitable for large-scale visual learning which is end-to-end trainable, and shows such models have distinct advantages over state-of-the-art models for recognition or generation which are separately defined and/or optimized.
Abstract: Models based on deep convolutional networks have dominated recent image interpretation tasks; we investigate whether models which are also recurrent, or “temporally deep”, are effective for tasks involving sequences, visual and otherwise. We develop a novel recurrent convolutional architecture suitable for large-scale visual learning which is end-to-end trainable, and demonstrate the value of these models on benchmark video recognition tasks, image description and retrieval problems, and video narration challenges. In contrast to current models which assume a fixed spatio-temporal receptive field or simple temporal averaging for sequential processing, recurrent convolutional models are “doubly deep” in that they can be compositional in spatial and temporal “layers”. Such models may have advantages when target concepts are complex and/or training data are limited. Learning long-term dependencies is possible when nonlinearities are incorporated into the network state updates. Long-term RNN models are appealing in that they directly can map variable-length inputs (e.g., video frames) to variable length outputs (e.g., natural language text) and can model complex temporal dynamics; yet they can be optimized with backpropagation. Our recurrent long-term models are directly connected to modern visual convnet models and can be jointly trained to simultaneously learn temporal dynamics and convolutional perceptual representations. Our results show such models have distinct advantages over state-of-the-art models for recognition or generation which are separately defined and/or optimized.
4,206 citations
Posted Content•
TL;DR: A novel recurrent convolutional architecture suitable for large-scale visual learning which is end-to-end trainable, and shows such models have distinct advantages over state-of-the-art models for recognition or generation which are separately defined and/or optimized.
Abstract: Models based on deep convolutional networks have dominated recent image interpretation tasks; we investigate whether models which are also recurrent, or "temporally deep", are effective for tasks involving sequences, visual and otherwise. We develop a novel recurrent convolutional architecture suitable for large-scale visual learning which is end-to-end trainable, and demonstrate the value of these models on benchmark video recognition tasks, image description and retrieval problems, and video narration challenges. In contrast to current models which assume a fixed spatio-temporal receptive field or simple temporal averaging for sequential processing, recurrent convolutional models are "doubly deep"' in that they can be compositional in spatial and temporal "layers". Such models may have advantages when target concepts are complex and/or training data are limited. Learning long-term dependencies is possible when nonlinearities are incorporated into the network state updates. Long-term RNN models are appealing in that they directly can map variable-length inputs (e.g., video frames) to variable length outputs (e.g., natural language text) and can model complex temporal dynamics; yet they can be optimized with backpropagation. Our recurrent long-term models are directly connected to modern visual convnet models and can be jointly trained to simultaneously learn temporal dynamics and convolutional perceptual representations. Our results show such models have distinct advantages over state-of-the-art models for recognition or generation which are separately defined and/or optimized.
3,935 citations
Cites methods from "YouTube2Text: Recognizing and Descr..."
...[11, 30, 17, 3, 6, 17, 41, 42] propose methods for generating sentence descriptions for video, but to our knowledge we present the first application of deep models to the video description task....
[...]
07 Dec 2015
TL;DR: The task of free-form and open-ended Visual Question Answering (VQA) is proposed, given an image and a natural language question about the image, the task is to provide an accurate natural language answer.
Abstract: We propose the task of free-form and open-ended Visual Question Answering (VQA). Given an image and a natural language question about the image, the task is to provide an accurate natural language answer. Mirroring real-world scenarios, such as helping the visually impaired, both the questions and answers are open-ended. Visual questions selectively target different areas of an image, including background details and underlying context. As a result, a system that succeeds at VQA typically needs a more detailed understanding of the image and complex reasoning than a system producing generic image captions. Moreover, VQA is amenable to automatic evaluation, since many open-ended answers contain only a few words or a closed set of answers that can be provided in a multiple-choice format. We provide a dataset containing ~0.25M images, ~0.76M questions, and ~10M answers (www.visualqa.org), and discuss the information it provides. Numerous baselines for VQA are provided and compared with human performance.
3,513 citations
Cites methods from "YouTube2Text: Recognizing and Descr..."
...Related to VQA are the tasks of image tagging [8, 23], image captioning [24, 14, 34, 7, 13, 45, 9, 19, 32, 21] and video captioning [39, 17], where words or sentences are generated to describe visual content....
[...]
Posted Content•
TL;DR: The task of free-form and open-ended Visual Question Answering (VQA) is proposed, given an image and a natural language question about the image, the task is to provide an accurate natural language answer.
Abstract: We propose the task of free-form and open-ended Visual Question Answering (VQA). Given an image and a natural language question about the image, the task is to provide an accurate natural language answer. Mirroring real-world scenarios, such as helping the visually impaired, both the questions and answers are open-ended. Visual questions selectively target different areas of an image, including background details and underlying context. As a result, a system that succeeds at VQA typically needs a more detailed understanding of the image and complex reasoning than a system producing generic image captions. Moreover, VQA is amenable to automatic evaluation, since many open-ended answers contain only a few words or a closed set of answers that can be provided in a multiple-choice format. We provide a dataset containing ~0.25M images, ~0.76M questions, and ~10M answers (this http URL), and discuss the information it provides. Numerous baselines and methods for VQA are provided and compared with human performance. Our VQA demo is available on CloudCV (this http URL).
2,365 citations
Cites methods from "YouTube2Text: Recognizing and Descr..."
...Related to VQA are the tasks of image tagging [8, 23], image captioning [24, 14, 34, 7, 13, 45, 9, 19, 32, 21] and video captioning [39, 17], where words or sentences are generated to describe visual content....
[...]
TL;DR: This paper surveys the recent advances in multimodal machine learning itself and presents them in a common taxonomy to enable researchers to better understand the state of the field and identify directions for future research.
Abstract: Our experience of the world is multimodal - we see objects, hear sounds, feel texture, smell odors, and taste flavors Modality refers to the way in which something happens or is experienced and a research problem is characterized as multimodal when it includes multiple such modalities In order for Artificial Intelligence to make progress in understanding the world around us, it needs to be able to interpret such multimodal signals together Multimodal machine learning aims to build models that can process and relate information from multiple modalities It is a vibrant multi-disciplinary field of increasing importance and with extraordinary potential Instead of focusing on specific multimodal applications, this paper surveys the recent advances in multimodal machine learning itself and presents them in a common taxonomy We go beyond the typical early and late fusion categorization and identify broader challenges that are faced by multimodal machine learning, namely: representation, translation, alignment, fusion, and co-learning This new taxonomy will enable researchers to better understand the state of the field and identify directions for future research
1,945 citations
References
More filters
TL;DR: Issues such as solving SVM optimization problems theoretical convergence multiclass classification probability estimates and parameter selection are discussed in detail.
Abstract: LIBSVM is a library for Support Vector Machines (SVMs). We have been actively developing this package since the year 2000. The goal is to help users to easily apply SVM to their applications. LIBSVM has gained wide popularity in machine learning and many other areas. In this article, we present all implementation details of LIBSVM. Issues such as solving SVM optimization problems theoretical convergence multiclass classification probability estimates and parameter selection are discussed in detail.
40,826 citations
"YouTube2Text: Recognizing and Descr..." refers methods in this paper
...…81.03 45.71 28.45 OU 92.57 46.83 46.66 93.72 61.19 58.41 Table 2: Comparison of WUP Similarity which combines the outputs of the SVMs according to the WordNet hierarchy and chooses the appropriate level of generalization by setting the accuracy to some prespecified value (0.9 in our experiments)....
[...]
...Then we combine activity and object descriptors using a multi-channel approach [31] and pass it to a non-linear SVM [3] (see Section 7)....
[...]
TL;DR: The state-of-the-art in evaluated methods for both classification and detection are reviewed, whether the methods are statistically different, what they are learning from the images, and what the methods find easy or confuse.
Abstract: The Pascal Visual Object Classes (VOC) challenge is a benchmark in visual object category recognition and detection, providing the vision and machine learning communities with a standard dataset of images and annotation, and standard evaluation procedures. Organised annually from 2005 to present, the challenge and its associated dataset has become accepted as the benchmark for object detection.
This paper describes the dataset and evaluation procedure. We review the state-of-the-art in evaluated methods for both classification and detection, analyse whether the methods are statistically different, what they are learning from the images (e.g. the object or its context), and what the methods find easy or confuse. The paper concludes with lessons learnt in the three year history of the challenge, and proposes directions for future improvement and extension.
15,935 citations
"YouTube2Text: Recognizing and Descr..." refers methods in this paper
...Subsets of this data were previously used by [20] and [14], however, contrary to these works, we use all the videos, and not only the 20 objects included in the PASCAL dataset [9]....
[...]
TL;DR: An object detection system based on mixtures of multiscale deformable part models that is able to represent highly variable object classes and achieves state-of-the-art results in the PASCAL object detection challenges is described.
Abstract: We describe an object detection system based on mixtures of multiscale deformable part models. Our system is able to represent highly variable object classes and achieves state-of-the-art results in the PASCAL object detection challenges. While deformable part models have become quite popular, their value had not been demonstrated on difficult benchmarks such as the PASCAL data sets. Our system relies on new methods for discriminative training with partially labeled data. We combine a margin-sensitive approach for data-mining hard negative examples with a formalism we call latent SVM. A latent SVM is a reformulation of MI--SVM in terms of latent variables. A latent SVM is semiconvex, and the training problem becomes convex once latent information is specified for the positive examples. This leads to an iterative training algorithm that alternates between fixing latent values for positive examples and optimizing the latent SVM objective function.
10,501 citations
"YouTube2Text: Recognizing and Descr..." refers background or methods in this paper
...Second, for each video we extract 2 frames per second, and for each frame, apply the object detectors proposed in [11] and [18], and select the maximum score assigned to each object in any frame....
[...]
...The flat (FL) baseline predicts the most confident output for each SVM trained over the whole set of labels without any hierarchy or grouping....
[...]
01 Jan 1999
4,584 citations
"YouTube2Text: Recognizing and Descr..." refers methods in this paper
...This idea can be used to expand “coarse” activity detections, obtained by training classifiers on available (possibly limited) activity training data, with “finer” activities unseen at training time....
[...]
TL;DR: In this article, a large collection of images with ground truth labels is built to be used for object detection and recognition research, such data is useful for supervised learning and quantitative evaluation.
Abstract: We seek to build a large collection of images with ground truth labels to be used for object detection and recognition research. Such data is useful for supervised learning and quantitative evaluation. To achieve this, we developed a web-based tool that allows easy image annotation and instant sharing of such annotations. Using this annotation tool, we have collected a large dataset that spans many object categories, often containing multiple instances over a wide variety of images. We quantify the contents of the dataset and compare against existing state of the art datasets used for object recognition and detection. Also, we show how to extend the dataset to automatically enhance object labels with WordNet, discover object parts, recover a depth ordering of objects in a scene, and increase the number of labels using minimal user supervision and images from the web.
3,501 citations