In this paper we present a self-supervised method for representation learning utilizing two different modalities. Based on the observation that cross-modal information has a high semantic meaning we propose a method to effectively exploit this signal. For our approach we utilize video data since it is available on a large scale and provides easily accessible modalities given by RGB and optical flow. We demonstrate state-of-the-art performance on highly contested action recognition datasets in the context of self-supervised learning. We show that our feature representation also transfers to other tasks and conduct extensive ablation studies to validate our core contributions. Code and model can be found at https://github.com/nawidsayed/Cross-and-Learn.
We propose 'Hide-and-Seek' a general purpose data augmentation technique, which is complementary to existing data augmentation techniques and is beneficial for various visual recognition tasks. The key idea is to hide patches in a training image randomly, in order to force the network to seek other relevant content when the most discriminative content is hidden. Our approach only needs to modify the input image and can work with any network to improve its performance. During testing, it does not need to hide any patches. The main advantage of Hide-and-Seek over existing data augmentation techniques is its ability to improve object localization accuracy in the weakly-supervised setting, and we therefore use this task to motivate the approach. However, Hide-and-Seek is not tied only to the image localization task, and can generalize to other forms of visual input like videos, as well as other recognition tasks like image classification, temporal action localization, semantic segmentation, emotion recognition, age/gender estimation, and person re-identification. We perform extensive experiments to showcase the advantage of Hide-and-Seek on these various visual recognition problems.
We propose DeepGRU, a novel end-to-end deep network model informed by recent developments in deep learning for gesture and action recognition, that is streamlined and device-agnostic. DeepGRU, which uses only raw skeleton, pose or vector data is quickly understood, implemented, and trained, and yet achieves state-of-the-art results on challenging datasets. At the heart of our method lies a set of stacked gated recurrent units (GRU), two fully-connected layers and a novel global attention model. We evaluate our method on seven publicly available datasets, containing various number of samples and spanning over a broad range of interactions (full-body, multi-actor, hand gestures, etc.). In all but one case we outperform the state-of-the-art pose-based methods. For instance, we achieve a recognition accuracy of 84.9% and 92.3% on cross-subject and cross-view tests of the NTU RGB+D dataset respectively, and also 100% recognition accuracy on the UT-Kinect dataset. While DeepGRU works well on large datasets with many training samples, we show that even in the absence of a large number of training data, and with as little as four samples per class, DeepGRU can beat traditional methods specifically designed for small training sets. Lastly, we demonstrate that even without powerful hardware, and using only the CPU, our method can still be trained in under 10 minutes on small-scale datasets, making it an enticing choice for rapid application prototyping and development.
Visual data and text data are composed of information at multiple granularities. A video can describe a complex scene that is composed of multiple clips or shots, where each depicts a semantically coherent event or action. Similarly, a paragraph may contain sentences with different topics, which collectively conveys a coherent message or story. In this paper, we investigate the modeling techniques for such hierarchical sequential data where there are correspondences across multiple modalities. Specifically, we introduce hierarchical sequence embedding (HSE), a generic model for embedding sequential data of different modalities into hierarchically semantic spaces, with either explicit or implicit correspondence information. We perform empirical studies on large-scale video and paragraph retrieval datasets and demonstrated superior performance by the proposed methods. Furthermore, we examine the effectiveness of our learned embeddings when applied to downstream tasks. We show its utility in zero-shot action recognition and video captioning.
We build CSI-Net, a unified Deep Neural Network~(DNN), to learn the representation of WiFi signals. Using CSI-Net, we jointly solved two body characterization problems: biometrics estimation (including body fat, muscle, water, and bone rates) and person recognition. We also demonstrated the application of CSI-Net on two distinctive pose recognition tasks: the hand sign recognition (fine-scaled action of the hand) and falling detection (coarse-scaled motion of the body).
In this paper, we propose a convolutional layer inspired by optical flow algorithms to learn motion representations. Our representation flow layer is a fully-differentiable layer designed to capture the `flow' of any representation channel within a convolutional neural network for action recognition. Its parameters for iterative flow optimization are learned in an end-to-end fashion together with the other CNN model parameters, maximizing the action recognition performance. Furthermore, we newly introduce the concept of learning `flow of flow' representations by stacking multiple representation flow layers. We conducted extensive experimental evaluations, confirming its advantages over previous recognition models using traditional optical flows in both computational speed and performance. Code/models available here: https://piergiaj.github.io/rep-flow-site/
Human actions comprise of joint motion of articulated body parts or `gestures'. Human skeleton is intuitively represented as a sparse graph with joints as nodes and natural connections between them as edges. Graph convolutional networks have been used to recognize actions from skeletal videos. We introduce a part-based graph convolutional network (PB-GCN) for this task, inspired by Deformable Part-based Models (DPMs). We divide the skeleton graph into four subgraphs with joints shared across them and learn a recognition model using a part-based graph convolutional network. We show that such a model improves performance of recognition, compared to a model using entire skeleton graph. Instead of using 3D joint coordinates as node features, we show that using relative coordinates and temporal displacements boosts performance. Our model achieves state-of-the-art performance on two challenging benchmark datasets NTURGB+D and HDM05, for skeletal action recognition.
Currently, the most common motion representation for action recognition is optical flow. Optical flow is based on particle tracking which adheres to a Lagrangian perspective on dynamics. In contrast to the Lagrangian perspective, the Eulerian model of dynamics does not track, but describes local changes. For video, an Eulerian phase-based motion representation, using complex steerable filters, has been successfully employed recently for motion magnification and video frame interpolation. Inspired by these previous works, here, we proposes learning Eulerian motion representations in a deep architecture for action recognition. We learn filters in the complex domain in an end-to-end manner. We design these complex filters to resemble complex Gabor filters, typically employed for phase-information extraction. We propose a phase-information extraction module, based on these complex filters, that can be used in any network architecture for extracting Eulerian representations. We experimentally analyze the added value of Eulerian motion representations, as extracted by our proposed phase extraction module, and compare with existing motion representations based on optical flow, on the UCF101 dataset.
We introduce a method for learning adversarial perturbations targeted to individual images or videos. The learned perturbations are found to be sparse while at the same time containing a high level of feature detail. Thus, the extracted perturbations allow a form of object or action recognition and provide insights into what features the studied deep neural network models consider important when reaching their classification decisions. From an adversarial point of view, the sparse perturbations successfully confused the models into misclassifying, although the perturbed samples still belonged to the same original class by visual examination. This is discussed in terms of a prospective data augmentation scheme. The sparse yet high-quality perturbations may also be leveraged for image or video compression.
Many videos depict people, and it is their interactions that inform us of their activities, relation to one another and the cultural and social setting. With advances in human action recognition, researchers have begun to address the automated recognition of these human-human interactions from video. The main challenges stem from dealing with the considerable variation in recording setting, the appearance of the people depicted and the coordinated performance of their interaction. This survey provides a summary of these challenges and datasets to address these, followed by an in-depth discussion of relevant vision-based recognition and detection methods. We focus on recent, promising work based on deep learning and convolutional neural networks (CNNs). Finally, we outline directions to overcome the limitations of the current state-of-the-art to analyze and, eventually, understand social human actions.
Despite the recent progress in video understanding and the continuous rate of improvement in temporal action localization throughout the years, it is still unclear how far (or close?) we are to solving the problem. To this end, we introduce a new diagnostic tool to analyze the performance of temporal action detectors in videos and compare different methods beyond a single scalar metric. We exemplify the use of our tool by analyzing the performance of the top rewarded entries in the latest ActivityNet action localization challenge. Our analysis shows that the most impactful areas to work on are: strategies to better handle temporal context around the instances, improving the robustness w.r.t. the instance absolute and relative size, and strategies to reduce the localization errors. Moreover, our experimental analysis finds the lack of agreement among annotator is not a major roadblock to attain progress in the field. Our diagnostic tool is publicly available to keep fueling the minds of other researchers with additional insights about their algorithms.
This paper aims to improve privacy-preserving visual recognition, an increasingly demanded feature in smart camera applications, by formulating a unique adversarial training framework. The proposed framework explicitly learns a degradation transform for the original video inputs, in order to optimize the trade-off between target task performance and the associated privacy budgets on the degraded video. A notable challenge is that the privacy budget, often defined and measured in task-driven contexts, cannot be reliably indicated using any single model performance, because a strong protection of privacy has to sustain against any possible model that tries to hack privacy information. Such an uncommon situation has motivated us to propose two strategies, i.e., budget model restarting and ensemble, to enhance the generalization of the learned degradation on protecting privacy against unseen hacker models. Novel training strategies, evaluation protocols, and result visualization methods have been designed accordingly. Two experiments on privacy-preserving action recognition, with privacy budgets defined in various ways, manifest the compelling effectiveness of the proposed framework in simultaneously maintaining high target task (action recognition) performance while suppressing the privacy breach risk.
Temporal Action Localization (TAL) in untrimmed video is important for many applications. But it is very expensive to annotate the segment-level ground truth (action class and temporal boundary). This raises the interest of addressing TAL with weak supervision, namely only video-level annotations are available during training). However, the state-of-the-art weakly-supervised TAL methods only focus on generating good Class Activation Sequence (CAS) over time but conduct simple thresholding on CAS to localize actions. In this paper, we first develop a novel weakly-supervised TAL framework called AutoLoc to directly predict the temporal boundary of each action instance. We propose a novel Outer-Inner-Contrastive (OIC) loss to automatically discover the needed segment-level supervision for training such a boundary predictor. Our method achieves dramatically improved performance: under the IoU threshold 0.5, our method improves mAP on THUMOS'14 from 13.7% to 21.2% and mAP on ActivityNet from 7.4% to 27.3%. It is also very encouraging to see that our weakly-supervised method achieves comparable results with some fully-supervised methods.
We propose a Variational Time Series Feature Extractor (VTSFE), inspired by the VAE-DMP model of Chen et al., to be used for action recognition and prediction. Our method is based on variational autoencoders. It improves VAE-DMP in that it has a better noise inference model, a simpler transition model constraining the acceleration in the trajectories of the latent space, and a tighter lower bound for the variational inference. We apply the method for classification and prediction of whole-body movements on a dataset with 7 tasks and 10 demonstrations per task, recorded with a wearable motion capture suit. The comparison with VAE and VAE-DMP suggests the better performance of our method for feature extraction. An open-source software implementation of each method with TensorFlow is also provided. In addition, a more detailed version of this work can be found in the indicated code repository. Although it was meant to, the VTSFE hasn't been tested for action prediction, due to a lack of time in the context of Maxime Chaveroche's Master thesis at INRIA.
Dynamic imaging is a recently proposed action description paradigm for simultaneously capturing motion and temporal evolution information, particularly in the context of deep convolutional neural networks (CNNs). Compared with optical flow for motion characterization, dynamic imaging exhibits superior efficiency and compactness. Inspired by the success of dynamic imaging in RGB video, this study extends it to the depth domain. To better exploit three-dimensional (3D) characteristics, multi-view dynamic images are proposed. In particular, the raw depth video is densely projected with respect to different virtual imaging viewpoints by rotating the virtual camera within the 3D space. Subsequently, dynamic images are extracted from the obtained multi-view depth videos and multi-view dynamic images are thus constructed from these images. Accordingly, more view-tolerant visual cues can be involved. A novel CNN model is then proposed to perform feature learning on multi-view dynamic images. Particularly, the dynamic images from different views share the same convolutional layers but correspond to different fully connected layers. This is aimed at enhancing the tuning effectiveness on shallow convolutional layers by alleviating the gradient vanishing problem. Moreover, as the spatial occurrence variation of the actions may impair the CNN, an action proposal approach is also put forth. In experiments, the proposed approach can achieve state-of-the-art performance on three challenging datasets.
We present a method to learn a joint multimodal representation space that enables recognition of unseen activities in videos. We first compare the effect of placing various constraints on the embedding space using paired text and video data. We also propose a method to improve the joint embedding space using an adversarial formulation, allowing it to benefit from unpaired text and video data. By using unpaired text data, we show the ability to learn a representation that better captures unseen activities. In addition to testing on publicly available datasets, we introduce a new, large-scale text/video dataset. We experimentally confirm that using paired and unpaired data to learn a shared embedding space benefits three difficult tasks (i) zero-shot activity classification, (ii) unsupervised activity discovery, and (iii) unseen activity captioning, outperforming the state-of-the-arts.
Diverse input data modalities can provide complementary cues for several tasks, usually leading to more robust algorithms and better performance. However, while a (training) dataset could be accurately designed to include a variety of sensory inputs, it is often the case that not all modalities could be available in real life (testing) scenarios, where a model has to be deployed. This raises the challenge of how to learn robust representations leveraging multimodal data in the training stage, while considering limitations at test time, such as noisy or missing modalities. This paper presents a new approach for multimodal video action recognition, developed within the unified frameworks of distillation and privileged information, named generalized distillation. Particularly, we consider the case of learning representations from depth and RGB videos, while relying on RGB data only at test time. We propose a new approach to train an hallucination network that learns to distill depth features through multiplicative connections of spatiotemporal representations, leveraging soft labels and hard labels, as well as distance between feature maps. We report state-of-the-art results on video action classification on the largest multimodal dataset available for this task, the NTU RGB+D. Code available at https://github.com/ncgarcia/modality-distillation .
We investigate architectures of discriminatively trained deep Convolutional Networks (ConvNets) for action recognition in video. The challenge is to capture the complementary information on appearance from still frames and motion between frames. We also aim to generalise the best performing hand-crafted features within a data-driven learning framework. Our contribution is three-fold. First, we propose a two-stream ConvNet architecture which incorporates spatial and temporal networks. Second, we demonstrate that a ConvNet trained on multi-frame dense optical flow is able to achieve very good performance in spite of limited training data. Finally, we show that multi-task learning, applied to two different action classification datasets, can be used to increase the amount of training data and improve the performance on both. Our architecture is trained and evaluated on the standard video actions benchmarks of UCF-101 and HMDB-51, where it is competitive with the state of the art. It also exceeds by a large margin previous attempts to use deep nets for video classification.
Deep convolutional networks have achieved great success for object recognition in still images. However, for action recognition in videos, the improvement of deep convolutional networks is not so evident. We argue that there are two reasons that could probably explain this result. First the current network architectures (e.g. Two-stream ConvNets) are relatively shallow compared with those very deep models in image domain (e.g. VGGNet, GoogLeNet), and therefore their modeling capacity is constrained by their depth. Second, probably more importantly, the training dataset of action recognition is extremely small compared with the ImageNet dataset, and thus it will be easy to over-fit on the training dataset. To address these issues, this report presents very deep two-stream ConvNets for action recognition, by adapting recent very deep architectures into video domain. However, this extension is not easy as the size of action recognition is quite small. We design several good practices for the training of very deep two-stream ConvNets, namely (i) pre-training for both spatial and temporal nets, (ii) smaller learning rates, (iii) more data augmentation techniques, (iv) high drop out ratio. Meanwhile, we extend the Caffe toolbox into Multi-GPU implementation with high computational efficiency and low memory consumption. We verify the performance of very deep two-stream ConvNets on the dataset of UCF101 and it achieves the recognition accuracy of $91.4\%$.
We have implemented a convolutional neural network designed for processing sparse three-dimensional input data. The world we live in is three dimensional so there are a large number of potential applications including 3D object recognition and analysis of space-time objects. In the quest for efficiency, we experiment with CNNs on the 2D triangular-lattice and 3D tetrahedral-lattice.