Videos are a rich source of high-dimensional structured data, with a wide range of interacting components at varying levels of granularity. In order to improve understanding of unconstrained internet videos, it is important to consider the role of labels at separate levels of abstraction. In this paper, we consider the use of the Bidirectional Inference Neural Network (BINN) for performing graph-based inference in label space for the task of video classification. We take advantage of the inherent hierarchy between labels at increasing granularity. The BINN is evaluated on the first and second release of the YouTube-8M large scale multilabel video dataset. Our results demonstrate the effectiveness of BINN, achieving significant improvements against baseline models.
Visual data such as images and videos contain a rich source of structured semantic labels as well as a wide range of interacting components. Visual content could be assigned with fine-grained labels describing major components, coarse-grained labels depicting high level abstractions, or a set of labels revealing attributes. Such categorization over different, interacting layers of labels evinces the potential for a graph-based encoding of label information. In this paper, we exploit this rich structure for performing graph-based inference in label space for a number of tasks: multi-label image and video classification and action detection in untrimmed videos. We consider the use of the Bidirectional Inference Neural Network (BINN) and Structured Inference Neural Network (SINN) for performing graph-based inference in label space and propose a Long Short-Term Memory (LSTM) based extension for exploiting activity progression on untrimmed videos. The methods were evaluated on (i) the Animal with Attributes (AwA), Scene Understanding (SUN) and NUS-WIDE datasets for multi-label image classification, (ii) the first two releases of the YouTube-8M large scale dataset for multi-label video classification, and (iii) the THUMOS'14 and MultiTHUMOS video datasets for action detection. Our results demonstrate the effectiveness of structured label inference in these challenging tasks, achieving significant improvements against baselines.
Accelerated by the tremendous increase in Internet bandwidth and storage space, video data has been generated, published and spread explosively, becoming an indispensable part of today's big data. In this paper, we focus on reviewing two lines of research aiming to stimulate the comprehension of videos with deep learning: video classification and video captioning. While video classification concentrates on automatically labeling video clips based on their semantic contents like human actions or complex events, video captioning attempts to generate a complete and natural sentence, enriching the single label as in video classification, to capture the most informative dynamics in videos. In addition, we also provide a review of popular benchmarks and competitions, which are critical for evaluating the technical progress of this vibrant field.
Current methods for video analysis often extract frame-level features using pre-trained convolutional neural networks (CNNs). Such features are then aggregated over time e.g., by simple temporal averaging or more sophisticated recurrent neural networks such as long short-term memory (LSTM) or gated recurrent units (GRU). In this work we revise existing video representations and study alternative methods for temporal aggregation. We first explore clustering-based aggregation layers and propose a two-stream architecture aggregating audio and visual features. We then introduce a learnable non-linear unit, named Context Gating, aiming to model interdependencies among network activations. Our experimental results show the advantage of both improvements for the task of video classification. In particular, we evaluate our method on the large-scale multi-modal Youtube-8M v2 dataset and outperform all other methods in the Youtube 8M Large-Scale Video Understanding challenge.
In this paper, a novel video classification method is presented that aims to recognize different categories of third-person videos efficiently. Our motivation is to achieve a light model that could be trained with insufficient training data. With this intuition, the processing of the 3-dimensional video input is broken to 1D in temporal dimension on top of the 2D in spatial. The processes related to 2D spatial frames are being done by utilizing pre-trained networks with no training phase. The only step which involves training is to classify the 1D time series resulted from the description of the 2D signals. As a matter of fact, optical flow images are first calculated from consecutive frames and described by pre-trained CNN networks. Their dimension is then reduced using PCA. By stacking the description vectors beside each other, a multi-channel time series is created for each video. Each channel of the time series represents a specific feature and follows it over time. The main focus of the proposed method is to classify the obtained time series effectively. Towards this, the idea is to let the machine learn temporal features. This is done by training a multi-channel one dimensional Convolutional Neural Network (1D-CNN). The 1D-CNN learns the features along the only temporal dimension. Hence, the number of training parameters decreases significantly which would result in the trainability of the method on even smaller datasets. It is illustrated that the proposed method could reach the state-of-the-art results on two public datasets UCF11, jHMDB and competitive results on HMDB51.
MLtuner automatically tunes settings for training tunables (such as the learning rate, the momentum, the mini-batch size, and the data staleness bound) that have a significant impact on large-scale machine learning (ML) performance. Traditionally, these tunables are set manually, which is unsurprisingly error-prone and difficult to do without extensive domain knowledge. MLtuner uses efficient snapshotting, branching, and optimization-guided online trial-and-error to find good initial settings as well as to re-tune settings during execution. Experiments show that MLtuner can robustly find and re-tune tunable settings for a variety of ML applications, including image classification (for 3 models and 2 datasets), video classification, and matrix factorization. Compared to state-of-the-art ML auto-tuning approaches, MLtuner is more robust for large problems and over an order of magnitude faster.
Human actions often involve complex interactions across several inter-related objects in the scene. However, existing approaches to fine-grained video understanding or visual relationship detection often rely on single object representation or pairwise object relationships. Furthermore, learning interactions across multiple objects in hundreds of frames for video is computationally infeasible and performance may suffer since a large combinatorial space has to be modeled. In this paper, we propose to efficiently learn higher-order interactions between arbitrary subgroups of objects for fine-grained video understanding. We demonstrate that modeling object interactions significantly improves accuracy for both action recognition and video captioning, while saving more than 3-times the computation over traditional pairwise relationships. The proposed method is validated on two large-scale datasets: Kinetics and ActivityNet Captions. Our SINet and SINet-Caption achieve state-of-the-art performances on both datasets even though the videos are sampled at a maximum of 1 FPS. To the best of our knowledge, this is the first work modeling object interactions on open domain large-scale video datasets, and we additionally model higher-order object interactions which improves the performance with low computational costs.
In this paper, a new video classification methodology is proposed which can be applied in both first and third person videos. The main idea behind the proposed strategy is to capture complementary information of appearance and motion efficiently by performing two independent streams on the videos. The first stream is aimed to capture long-term motions from shorter ones by keeping track of how elements in optical flow images have changed over time. Optical flow images are described by pre-trained networks that have been trained on large scale image datasets. A set of multi-channel time series are obtained by aligning descriptions beside each other. For extracting motion features from these time series, PoT representation method plus a novel pooling operator is followed due to several advantages. The second stream is accomplished to extract appearance features which are vital in the case of video classification. The proposed method has been evaluated on both first and third-person datasets and results present that the proposed methodology reaches the state of the art successfully.
In this paper, we introduce a challenging new dataset, MLB-YouTube, designed for fine-grained activity detection. The dataset contains two settings: segmented video classification as well as activity detection in continuous videos. We experimentally compare various recognition approaches capturing temporal structure in activity videos, by classifying segmented videos and extending those approaches to continuous videos. We also compare models on the extremely difficult task of predicting pitch speed and pitch type from broadcast baseball videos. We find that learning temporal structure is valuable for fine-grained activity recognition.
We introduce DAiSEE, the first multi-label video classification dataset comprising of 9068 video snippets captured from 112 users for recognizing the user affective states of boredom, confusion, engagement, and frustration in the wild. The dataset has four levels of labels namely - very low, low, high, and very high for each of the affective states, which are crowd annotated and correlated with a gold standard annotation created using a team of expert psychologists. We have also established benchmark results on this dataset using state-of-the-art video classification methods that are available today. We believe that DAiSEE will provide the research community with challenges in feature extraction, context-based inference, and development of suitable machine learning methods for related tasks, thus providing a springboard for further research. The dataset is available for download at https://people.iith.ac.in/vineethnb/resources/daisee/index.html.
Both convolutional and recurrent operations are building blocks that process one local neighborhood at a time. In this paper, we present non-local operations as a generic family of building blocks for capturing long-range dependencies. Inspired by the classical non-local means method in computer vision, our non-local operation computes the response at a position as a weighted sum of the features at all positions. This building block can be plugged into many computer vision architectures. On the task of video classification, even without any bells and whistles, our non-local models can compete or outperform current competition winners on both Kinetics and Charades datasets. In static image recognition, our non-local models improve object detection/segmentation and pose estimation on the COCO suite of tasks. Code is available at https://github.com/facebookresearch/video-nonlocal-net .
Graph classification is a fundamental but challenging issue for numerous real-world applications. Despite recent great progress in image/video classification, convolutional neural networks (CNNs) cannot yet cater to graphs well because of graphical non-Euclidean topology. In this work, we propose a walk-steered convolutional (WSC) network to assemble the essential success of standard convolutional neural networks as well as the powerful representation ability of random walk. Instead of deterministic neighbor searching used in previous graphical CNNs, we construct multi-scale walk fields (a.k.a. local receptive fields) with random walk paths to depict subgraph structures and advocate graph scalability. To express the internal variations of a walk field, Gaussian mixture models are introduced to encode principal components of walk paths therein. As an analogy to a standard convolution kernel on image, Gaussian models implicitly coordinate those unordered vertices/nodes and edges in a local receptive field after projecting to the gradient space of Gaussian parameters. We further stack graph coarsening upon Gaussian encoding by using dynamic clustering, such that high-level semantics of graph can be well learned like the conventional pooling on image. The experimental results on several public datasets demonstrate the superiority of our proposed WSC method over many state-of-the-arts for graph classification.
We describe a DNN for video classification and captioning, trained end-to-end, with shared features, to solve tasks at different levels of granularity, exploring the link between granularity in a source task and the quality of learned features for transfer learning. For solving the new task domain in transfer learning, we freeze the trained encoder and fine-tune a neural net on the target domain. We train on the Something-Something dataset with over 220, 000 videos, and multiple levels of target granularity, including 50 action groups, 174 fine-grained action categories and captions. Classification and captioning with Something-Something are challenging because of the subtle differences between actions, applied to thousands of different object classes, and the diversity of captions penned by crowd actors. Our model performs better than existing classification baselines for SomethingSomething, with impressive fine-grained results. And it yields a strong baseline on the new Something-Something captioning task. Experiments reveal that training with more fine-grained tasks tends to produce better features for transfer learning.
Zero-Shot Learning (ZSL) in video classification is a promising research direction, which aims to tackle the challenge from explosive growth of video categories. Most existing methods exploit seen-to-unseen correlation via learning a projection between visual and semantic spaces. However, such projection-based paradigms cannot fully utilize the discriminative information implied in data distribution, and commonly suffer from the information degradation issue caused by "heterogeneity gap". In this paper, we propose a visual data synthesis framework via GAN to address these problems. Specifically, both semantic knowledge and visual distribution are leveraged to synthesize video feature of unseen categories, and ZSL can be turned into typical supervised problem with the synthetic features. First, we propose multi-level semantic inference to boost video feature synthesis, which captures the discriminative information implied in joint visual-semantic distribution via feature-level and label-level semantic inference. Second, we propose Matching-aware Mutual Information Correlation to overcome information degradation issue, which captures seen-to-unseen correlation in matched and mismatched visual-semantic pairs by mutual information, providing the zero-shot synthesis procedure with robust guidance signals. Experimental results on four video datasets demonstrate that our approach can improve the zero-shot video classification performance significantly.
Video representation learning is a vital problem for classification task. Recently, a promising unsupervised paradigm termed self-supervised learning has emerged, which explores inherent supervisory signals implied in massive data for feature learning via solving auxiliary tasks. However, existing methods in this regard suffer from two limitations when extended to video classification. First, they focus only on a single task, whereas ignoring complementarity among different task-specific features and thus resulting in suboptimal video representation. Second, high computational and memory cost hinders their application in real-world scenarios. In this paper, we propose a graph-based distillation framework to address these problems: (1) We propose logits graph and representation graph to transfer knowledge from multiple self-supervised tasks, where the former distills classifier-level knowledge by solving a multi-distribution joint matching problem, and the latter distills internal feature knowledge from pairwise ensembled representations with tackling the challenge of heterogeneity among different features; (2) The proposal that adopts a teacher-student framework can reduce the redundancy of knowledge learnt from teachers dramatically, leading to a lighter student model that solves classification task more efficiently. Experimental results on 3 video datasets validate that our proposal not only helps learn better video representation but also compress model for faster inference.
In Actor and Observer we introduced a dataset linking the first and third-person video understanding domains, the Charades-Ego Dataset. In this paper we describe the egocentric aspect of the dataset and present annotations for Charades-Ego with 68,536 activity instances in 68.8 hours of first and third-person video, making it one of the largest and most diverse egocentric datasets available. Charades-Ego furthermore shares activity classes, scripts, and methodology with the Charades dataset, that consist of additional 82.3 hours of third-person video with 66,500 activity instances. Charades-Ego has temporal annotations and textual descriptions, making it suitable for egocentric video classification, localization, captioning, and new tasks utilizing the cross-modal nature of the data.
Spatiotemporal feature learning in videos is a fundamental problem in computer vision. This paper presents a new architecture, termed as Appearance-and-Relation Network (ARTNet), to learn video representation in an end-to-end manner. ARTNets are constructed by stacking multiple generic building blocks, called as SMART, whose goal is to simultaneously model appearance and relation from RGB input in a separate and explicit manner. Specifically, SMART blocks decouple the spatiotemporal learning module into an appearance branch for spatial modeling and a relation branch for temporal modeling. The appearance branch is implemented based on the linear combination of pixels or filter responses in each frame, while the relation branch is designed based on the multiplicative interactions between pixels or filter responses across multiple frames. We perform experiments on three action recognition benchmarks: Kinetics, UCF101, and HMDB51, demonstrating that SMART blocks obtain an evident improvement over 3D convolutions for spatiotemporal feature learning. Under the same training setting, ARTNets achieve superior performance on these three datasets to the existing state-of-the-art methods.
The state of the art in video understanding suffers from two problems: (1) The major part of reasoning is performed locally in the video, therefore, it misses important relationships within actions that span several seconds. (2) While there are local methods with fast per-frame processing, the processing of the whole video is not efficient and hampers fast video retrieval or online classification of long-term activities. In this paper, we introduce a network architecture that takes long-term content into account and enables fast per-video processing at the same time. The architecture is based on merging long-term content already in the network rather than in a post-hoc fusion. Together with a sampling strategy, which exploits that neighboring frames are largely redundant, this yields high-quality action classification and video captioning at up to 230 videos per second, where each video can consist of a few hundred frames. The approach achieves competitive performance across all datasets while being 10x to 80x faster than state-of-the-art methods.
Over the past few years, various tasks involving videos such as classification, description, summarization and question answering have received a lot of attention. Current models for these tasks compute an encoding of the video by treating it as a sequence of images and going over every image in the sequence. However, for longer videos this is very time consuming. In this paper, we focus on the task of video classification and aim to reduce the computational time by using the idea of distillation. Specifically, we first train a teacher network which looks at all the frames in a video and computes a representation for the video. We then train a student network whose objective is to process only a small fraction of the frames in the video and still produce a representation which is very close to the representation computed by the teacher network. This smaller student network involving fewer computations can then be employed at inference time for video classification. We experiment with the YouTube-8M dataset and show that the proposed student network can reduce the inference time by upto 30% with a very small drop in the performance