Human action or activity recognition in videos is a fundamental task in computer vision with applications in surveillance and monitoring, self-driving cars, sports analytics, human-robot interaction and many more. Traditional supervised methods require large annotated datasets for training, which are expensive and time-consuming to acquire. This work proposes a novel approach using Cross-Architecture Pseudo-Labeling with contrastive learning for semi-supervised action recognition. Our framework leverages both labeled and unlabelled data to robustly learn action representations in videos, combining pseudo-labeling with contrastive learning for effective learning from both types of samples. We introduce a novel cross-architecture approach where 3D Convolutional Neural Networks (3D CNNs) and video transformers (VIT) are utilised to capture different aspects of action representations; hence we call it ActNetFormer. The 3D CNNs excel at capturing spatial features and local dependencies in the temporal domain, while VIT excels at capturing long-range dependencies across frames. By integrating these complementary architectures within the ActNetFormer framework, our approach can effectively capture both local and global contextual information of an action. This comprehensive representation learning enables the model to achieve better performance in semi-supervised action recognition tasks by leveraging the strengths of each of these architectures. Experimental results on standard action recognition datasets demonstrate that our approach performs better than the existing methods, achieving state-of-the-art performance with only a fraction of labeled data. The official website of this work is available at: https://github.com/rana2149/ActNetFormer.
In this paper, we efficiently transfer the surpassing representation power of the vision foundation models, such as ViT and Swin, for video understanding with only a few trainable parameters. Previous adaptation methods have simultaneously considered spatial and temporal modeling with a unified learnable module but still suffered from fully leveraging the representative capabilities of image transformers. We argue that the popular dual-path (two-stream) architecture in video models can mitigate this problem. We propose a novel DualPath adaptation separated into spatial and temporal adaptation paths, where a lightweight bottleneck adapter is employed in each transformer block. Especially for temporal dynamic modeling, we incorporate consecutive frames into a grid-like frameset to precisely imitate vision transformers' capability that extrapolates relationships between tokens. In addition, we extensively investigate the multiple baselines from a unified perspective in video understanding and compare them with DualPath. Experimental results on four action recognition benchmarks prove that pretrained image transformers with DualPath can be effectively generalized beyond the data domain.
We present a learning algorithm for human activity recognition in videos. Our approach is designed for UAV videos, which are mainly acquired from obliquely placed dynamic cameras that contain a human actor along with background motion. Typically, the human actors occupy less than one-tenth of the spatial resolution. Our approach simultaneously harnesses the benefits of frequency domain representations, a classical analysis tool in signal processing, and data driven neural networks. We build a differentiable static-dynamic frequency mask prior to model the salient static and dynamic pixels in the video, crucial for the underlying task of action recognition. We use this differentiable mask prior to enable the neural network to intrinsically learn disentangled feature representations via an identity loss function. Our formulation empowers the network to inherently compute disentangled salient features within its layers. Further, we propose a cost-function encapsulating temporal relevance and spatial content to sample the most important frame within uniformly spaced video segments. We conduct extensive experiments on the UAV Human dataset and the NEC Drone dataset and demonstrate relative improvements of 5.72% - 13.00% over the state-of-the-art and 14.28% - 38.05% over the corresponding baseline model.
Human activity recognition is an emerging and important area in computer vision which seeks to determine the activity an individual or group of individuals are performing. The applications of this field ranges from generating highlight videos in sports, to intelligent surveillance and gesture recognition. Most activity recognition systems rely on a combination of convolutional neural networks (CNNs) to perform feature extraction from the data and recurrent neural networks (RNNs) to determine the time dependent nature of the data. This paper proposes and designs two transformer neural networks for human activity recognition: a recurrent transformer (ReT), a specialized neural network used to make predictions on sequences of data, as well as a vision transformer (ViT), a transformer optimized for extracting salient features from images, to improve speed and scalability of activity recognition. We have provided an extensive comparison of the proposed transformer neural networks with the contemporary CNN and RNN-based human activity recognition models in terms of speed and accuracy.
We propose the use of fractals as a means of efficient data augmentation. Specifically, we employ plasma fractals for adapting global image augmentation transformations into continuous local transforms. We formulate the diamond square algorithm as a cascade of simple convolution operations allowing efficient computation of plasma fractals on the GPU. We present the TorMentor image augmentation framework that is totally modular and deterministic across images and point-clouds. All image augmentation operations can be combined through pipelining and random branching to form flow networks of arbitrary width and depth. We demonstrate the efficiency of the proposed approach with experiments on document image segmentation (binarization) with the DIBCO datasets. The proposed approach demonstrates superior performance to traditional image augmentation techniques. Finally, we use extended synthetic binary text images in a self-supervision regiment and outperform the same model when trained with limited data and simple extensions.
Human activity recognition in videos is a challenging problem that has drawn a lot of interest, particularly when the goal requires the analysis of a large video database. AOLME project provides a collaborative learning environment for middle school students to explore mathematics, computer science, and engineering by processing digital images and videos. As part of this project, around 2200 hours of video data was collected for analysis. Because of the size of the dataset, it is hard to analyze all the videos of the dataset manually. Thus, there is a huge need for reliable computer-based methods that can detect activities of interest. My thesis is focused on the development of accurate methods for detecting and tracking objects in long videos. All the models are validated on videos from 7 different sessions, ranging from 45 minutes to 90 minutes. The keyboard detector achieved a very high average precision (AP) of 92% at 0.5 intersection over union (IoU). Furthermore, a combined system of the detector with a fast tracker KCF (159fps) was developed so that the algorithm runs significantly faster without sacrificing accuracy. For a video of 23 minutes having resolution 858X480 @ 30 fps, the detection alone runs at 4.7Xthe real-time, and the combined algorithm runs at 21Xthe real-time for an average IoU of 0.84 and 0.82, respectively. The hand detector achieved average precision (AP) of 72% at 0.5 IoU. The detection results were improved to 81% using optimal data augmentation parameters. The hand detector runs at 4.7Xthe real-time with AP of 81% at 0.5 IoU. The hand detection method was integrated with projections and clustering for accurate proposal generation. This approach reduced the number of false-positive hand detections by 80%. The overall hand detection system runs at 4Xthe real-time, capturing all the activity regions of the current collaborative group.
This paper presents an approach to perform human activity recognition in videos through the employment of a deep recurrent network, taking as inputs appearance and optical flow information. Our method proposes a novel architecture named BubbleNET, which is based on a recurrent layer dispersed into several modules (referred to as bubbles) along with an attention mechanism based on squeeze-and-excitation strategy, responsible to modulate each bubble contribution. Thereby, we intend to gather information from fundamentally correlated segments of the input data, creating a signature of components that characterize each activity. Our experiments, conducted on widely employed activity recognition datasets, support the existence of these signatures, evidenced by maps of bubble activations for every class of the datasets. To compare the approach to literature methods, mean accuracy is taken into account, for which BubbleNET obtained 97.62%, 91.70% and 82.60% on UCF-101, YUP++ and HMDB-51 datasets, respectively, being placed among state-of-the-art methods.
Human activity recognition in videos has been widely studied and has recently gained significant advances with deep learning approaches; however, it remains a challenging task. In this paper, we propose a novel framework that simultaneously considers both implicit and explicit representations of human interactions by fusing information of local image where the interaction actively occurred, primitive motion with the posture of individual subject's body parts, and the co-occurrence of overall appearance change. Human interactions change, depending on how the body parts of each human interact with the other. The proposed method captures the subtle difference between different interactions using interacting body part attention. Semantically important body parts that interact with other objects are given more weight during feature representation. The combined feature of interacting body part attention-based individual representation and the co-occurrence descriptor of the full-body appearance change is fed into long short-term memory to model the temporal dynamics over time in a single framework. We validate the effectiveness of the proposed method using four widely used public datasets by outperforming the competing state-of-the-art method.
Spiking neural networks (SNNs) can be used in low-power and embedded systems (such as emerging neuromorphic chips) due to their event-based nature. Also, they have the advantage of low computation cost in contrast to conventional artificial neural networks (ANNs), while preserving ANN's properties. However, temporal coding in layers of convolutional spiking neural networks and other types of SNNs has yet to be studied. In this paper, we provide insight into spatio-temporal feature extraction of convolutional SNNs in experiments designed to exploit this property. The shallow convolutional SNN outperforms state-of-the-art spatio-temporal feature extractor methods such as C3D, ConvLstm, and similar networks. Furthermore, we present a new deep spiking architecture to tackle real-world problems (in particular classification tasks) which achieved superior performance compared to other SNN methods on NMNIST (99.6%), DVS-CIFAR10 (69.2%) and DVS-Gesture (96.7%) and ANN methods on UCF-101 (42.1%) and HMDB-51 (21.5%) datasets. It is also worth noting that the training process is implemented based on variation of spatio-temporal backpropagation explained in the paper.
In this paper, we present an end-to-end future-prediction model that focuses on pedestrian safety. Specifically, our model uses previous video frames, recorded from the perspective of the vehicle, to predict if a pedestrian will cross in front of the vehicle. The long term goal of this work is to design a fully autonomous system that acts and reacts as a defensive human driver would --- predicting future events and reacting to mitigate risk. We focus on pedestrian-vehicle interactions because of the high risk of harm to the pedestrian if their actions are miss-predicted. Our end-to-end model consists of two stages: the first stage is an encoder/decoder network that learns to predict future video frames. The second stage is a deep spatio-temporal network that utilizes the predicted frames of the first stage to predict the pedestrian's future action. Our system achieves state-of-the-art accuracy on pedestrian behavior prediction and future frames prediction on the Joint Attention for Autonomous Driving (JAAD) dataset.
Current fully-supervised video datasets consist of only a few hundred thousand videos and fewer than a thousand domain-specific labels. This hinders the progress towards advanced video architectures. This paper presents an in-depth study of using large volumes of web videos for pre-training video models for the task of action recognition. Our primary empirical finding is that pre-training at a very large scale (over 65 million videos), despite on noisy social-media videos and hashtags, substantially improves the state-of-the-art on three challenging public action recognition datasets. Further, we examine three questions in the construction of weakly-supervised video action datasets. First, given that actions involve interactions with objects, how should one construct a verb-object pre-training label space to benefit transfer learning the most? Second, frame-based models perform quite well on action recognition; is pre-training for good image features sufficient or is pre-training for spatio-temporal features valuable for optimal transfer learning? Finally, actions are generally less well-localized in long videos vs. short videos; since action labels are provided at a video level, how should one choose video clips for best performance, given some fixed budget of number or minutes of videos?
Activity recognition in videos in a deep-learning setting---or otherwise---uses both static and pre-computed motion components. The method of combining the two components, whilst keeping the burden on the deep network less, still remains uninvestigated. Moreover, it is not clear what the level of contribution of individual components is, and how to control the contribution. In this work, we use a combination of CNN-generated static features and motion features in the form of motion tubes. We propose three schemas for combining static and motion components: based on a variance ratio, principal components, and Cholesky decomposition. The Cholesky decomposition based method allows the control of contributions. The ratio given by variance analysis of static and motion features match well with the experimental optimal ratio used in the Cholesky decomposition based method. The resulting activity recognition system is better or on par with existing state-of-the-art when tested with three popular datasets. The findings also enable us to characterize a dataset with respect to its richness in motion information.
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/
In this paper, a new network-transmission-based (NTB) algorithm is proposed for human activity recognition in videos. The proposed NTB algorithm models the entire scene as an error-free network. In this network, each node corresponds to a patch of the scene and each edge represents the activity correlation between the corresponding patches. Based on this network, we further model people in the scene as packages while human activities can be modeled as the process of package transmission in the network. By analyzing these specific "package transmission" processes, various activities can be effectively detected. The implementation of our NTB algorithm into abnormal activity detection and group activity recognition are described in detail in the paper. Experimental results demonstrate the effectiveness of our proposed algorithm.
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.
In this paper, we present a new feature representation for first-person videos. In first-person video understanding (e.g., activity recognition), it is very important to capture both entire scene dynamics (i.e., egomotion) and salient local motion observed in videos. We describe a representation framework based on time series pooling, which is designed to abstract short-term/long-term changes in feature descriptor elements. The idea is to keep track of how descriptor values are changing over time and summarize them to represent motion in the activity video. The framework is general, handling any types of per-frame feature descriptors including conventional motion descriptors like histogram of optical flows (HOF) as well as appearance descriptors from more recent convolutional neural networks (CNN). We experimentally confirm that our approach clearly outperforms previous feature representations including bag-of-visual-words and improved Fisher vector (IFV) when using identical underlying feature descriptors. We also confirm that our feature representation has superior performance to existing state-of-the-art features like local spatio-temporal features and Improved Trajectory Features (originally developed for 3rd-person videos) when handling first-person videos. Multiple first-person activity datasets were tested under various settings to confirm these findings.
Visual features are of vital importance for human action understanding in videos. This paper presents a new video representation, called trajectory-pooled deep-convolutional descriptor (TDD), which shares the merits of both hand-crafted features and deep-learned features. Specifically, we utilize deep architectures to learn discriminative convolutional feature maps, and conduct trajectory-constrained pooling to aggregate these convolutional features into effective descriptors. To enhance the robustness of TDDs, we design two normalization methods to transform convolutional feature maps, namely spatiotemporal normalization and channel normalization. The advantages of our features come from (i) TDDs are automatically learned and contain high discriminative capacity compared with those hand-crafted features; (ii) TDDs take account of the intrinsic characteristics of temporal dimension and introduce the strategies of trajectory-constrained sampling and pooling for aggregating deep-learned features. We conduct experiments on two challenging datasets: HMDB51 and UCF101. Experimental results show that TDDs outperform previous hand-crafted features and deep-learned features. Our method also achieves superior performance to the state of the art on these datasets (HMDB51 65.9%, UCF101 91.5%).
In this paper, we newly introduce the concept of temporal attention filters, and describe how they can be used for human activity recognition from videos. Many high-level activities are often composed of multiple temporal parts (e.g., sub-events) with different duration/speed, and our objective is to make the model explicitly learn such temporal structure using multiple attention filters and benefit from them. Our temporal filters are designed to be fully differentiable, allowing end-of-end training of the temporal filters together with the underlying frame-based or segment-based convolutional neural network architectures. This paper presents an approach of learning a set of optimal static temporal attention filters to be shared across different videos, and extends this approach to dynamically adjust attention filters per testing video using recurrent long short-term memory networks (LSTMs). This allows our temporal attention filters to learn latent sub-events specific to each activity. We experimentally confirm that the proposed concept of temporal attention filters benefits the activity recognition, and we visualize the learned latent sub-events.