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.

/pdf/long-term-recurrent-convolutional-networks-for-visual-3znyib4gwu.pdf

Long-term recurrent convolutional networks for visual recognition and description

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

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.

Long-term Recurrent Convolutional Networks for Visual Recognition and Description

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

Recently dense trajectories were shown to be an efficient video representation for action recognition and achieved state-of-the-art results on a variety of datasets. This paper improves their performance by taking into account camera motion to correct them. To estimate camera motion, we match feature points between frames using SURF descriptors and dense optical flow, which are shown to be complementary. These matches are, then, used to robustly estimate a homography with RANSAC. Human motion is in general different from camera motion and generates inconsistent matches. To improve the estimation, a human detector is employed to remove these matches. Given the estimated camera motion, we remove trajectories consistent with it. We also use this estimation to cancel out camera motion from the optical flow. This significantly improves motion-based descriptors, such as HOF and MBH. Experimental results on four challenging action datasets (i.e., Hollywood2, HMDB51, Olympic Sports and UCF50) significantly outperform the current state of the art.

/pdf/action-recognition-with-improved-trajectories-3z939q0pgz.pdf

Action Recognition with Improved Trajectories

Based on keypoints extracted as salient image patches, an image can be described as a "bag of visual words" and this representation has been used in scene classification. The choice of dimension, selection, and weighting of visual words in this representation is crucial to the classification performance but has not been thoroughly studied in previous work. Given the analogy between this representation and the bag-of-words representation of text documents, we apply techniques used in text categorization, including term weighting, stop word removal, feature selection, to generate image representations that differ in the dimension, selection, and weighting of visual words. The impact of these representation choices to scene classification is studied through extensive experiments on the TRECVID and PASCAL collection. This study provides an empirical basis for designing visual-word representations that are likely to produce superior classification performance.

http://www.cs.cmu.edu/~juny/Prof/papers/mir07jyang.pdf

Evaluating bag-of-visual-words representations in scene classification

Bag-of-features (BoF) deriving from local keypoints has recently appeared promising for object and scene classification. Whether BoF can naturally survive the challenges such as reliability and scalability of visual classification, nevertheless, remains uncertain due to various implementation choices. In this paper, we evaluate various factors which govern the performance of BoF. The factors include the choices of detector, kernel, vocabulary size and weighting scheme. We offer some practical insights in how to optimize the performance by choosing good keypoint detector and kernel. For the weighting scheme, we propose a novel soft-weighting method to assess the significance of a visual word to an image. We experimentally show that the proposed soft-weighting scheme can consistently offer better performance than other popular weighting methods. On both PASCAL-2005 and TRECVID-2006 datasets, our BoF setting generates competitive performance compared to the state-of-the-art techniques. We also show that the BoF is highly complementary to global features. By incorporating the BoF with color and texture features, an improvement of 50% is reported on TRECVID-2006 dataset.

/pdf/towards-optimal-bag-of-features-for-object-categorization-oejes79xvr.pdf

Towards optimal bag-of-features for object categorization and semantic video retrieval

We propose a unified approach for video summarization based on the analysis of video structures and video highlights. Two major components in our approach are scene modeling and highlight detection. Scene modeling is achieved by normalized cut algorithm and temporal graph analysis, while highlight detection is accomplished by motion attention modeling. In our proposed approach, a video is represented as a complete undirected graph and the normalized cut algorithm is carried out to globally and optimally partition the graph into video clusters. The resulting clusters form a directed temporal graph and a shortest path algorithm is proposed to efficiently detect video scenes. The attention values are then computed and attached to the scenes, clusters, shots, and subshots in a temporal graph. As a result, the temporal graph can inherently describe the evolution and perceptual importance of a video. In our application, video summaries that emphasize both content balance and perceptual quality can be generated directly from a temporal graph that embeds both the structure and attention information.

Video summarization and scene detection by graph modeling

Current web video search results rely exclusively on text keywords or user-supplied tags. A search on typical popular video often returns many duplicate and near-duplicate videos in the top results. This paper outlines ways to cluster and filter out the near-duplicate video using a hierarchical approach. Initial triage is performed using fast signatures derived from color histograms. Only when a video cannot be clearly classified as novel or near-duplicate using global signatures, we apply a more expensive local feature based near-duplicate detection which provides very accurate duplicate analysis through more costly computation. The results of 24 queries in a data set of 12,790 videos retrieved from Google, Yahoo! and YouTube show that this hierarchical approach can dramatically reduce redundant video displayed to the user in the top result set, at relatively small computational cost.

Practical elimination of near-duplicates from web video search

Retrieving recipes corresponding to given dish pictures facilitates the estimation of nutrition facts, which is crucial to various health relevant applications. The current approaches mostly focus on recognition of food category based on global dish appearance without explicit analysis of ingredient composition. Such approaches are incapable for retrieval of recipes with unknown food categories, a problem referred to as zero-shot retrieval. On the other hand, content-based retrieval without knowledge of food categories is also difficult to attain satisfactory performance due to large visual variations in food appearance and ingredient composition. As the number of ingredients is far less than food categories, understanding ingredients underlying dishes in principle is more scalable than recognizing every food category and thus is suitable for zero-shot retrieval. Nevertheless, ingredient recognition is a task far harder than food categorization, and this seriously challenges the feasibility of relying on them for retrieval. This paper proposes deep architectures for simultaneous learning of ingredient recognition and food categorization, by exploiting the mutual but also fuzzy relationship between them. The learnt deep features and semantic labels of ingredients are then innovatively applied for zero-shot retrieval of recipes. By experimenting on a large Chinese food dataset with images of highly complex dish appearance, this paper demonstrates the feasibility of ingredient recognition and sheds light on this zero-shot problem peculiar to cooking recipe retrieval.

Chong-Wah Ngo

Papers

Evaluating bag-of-visual-words representations in scene classification

Towards optimal bag-of-features for object categorization and semantic video retrieval

Video summarization and scene detection by graph modeling

Practical elimination of near-duplicates from web video search

Deep-based Ingredient Recognition for Cooking Recipe Retrieval