Online Temporal Action Localization (On-TAL) aims to immediately provide action instances from untrimmed streaming videos. The model is not allowed to utilize future frames and any processing techniques to modify past predictions, making On-TAL much more challenging. In this paper, we propose a simple yet effective framework, termed SimOn, that learns to predict action instances using the popular Transformer architecture in an end-to-end manner. Specifically, the model takes the current frame feature as a query and a set of past context information as keys and values of the Transformer. Different from the prior work that uses a set of outputs of the model as past contexts, we leverage the past visual context and the learnable context embedding for the current query. Experimental results on the THUMOS14 and ActivityNet1.3 datasets show that our model remarkably outperforms the previous methods, achieving a new state-of-the-art On-TAL performance. In addition, the evaluation for Online Detection of Action Start (ODAS) demonstrates the effectiveness and robustness of our method in the online setting. The code is available at https://github.com/TuanTNG/SimOn
Pixel space augmentation has grown in popularity in many Deep Learning areas, due to its effectiveness, simplicity, and low computational cost. Data augmentation for videos, however, still remains an under-explored research topic, as most works have been treating inputs as stacks of static images rather than temporally linked series of data. Recently, it has been shown that involving the time dimension when designing augmentations can be superior to its spatial-only variants for video action recognition. In this paper, we propose several novel enhancements to these techniques to strengthen the relationship between the spatial and temporal domains and achieve a deeper level of perturbations. The video action recognition results of our techniques outperform their respective variants in Top-1 and Top-5 settings on the UCF-101 and the HMDB-51 datasets.
Frozen pretrained models have become a viable alternative to the pretraining-then-finetuning paradigm for transfer learning. However, with frozen models there are relatively few parameters available for adapting to downstream tasks, which is problematic in computer vision where tasks vary significantly in input/output format and the type of information that is of value. In this paper, we present a study of frozen pretrained models when applied to diverse and representative computer vision tasks, including object detection, semantic segmentation and video action recognition. From this empirical analysis, our work answers the questions of what pretraining task fits best with this frozen setting, how to make the frozen setting more flexible to various downstream tasks, and the effect of larger model sizes. We additionally examine the upper bound of performance using a giant frozen pretrained model with 3 billion parameters (SwinV2-G) and find that it reaches competitive performance on a varied set of major benchmarks with only one shared frozen base network: 60.0 box mAP and 52.2 mask mAP on COCO object detection test-dev, 57.6 val mIoU on ADE20K semantic segmentation, and 81.7 top-1 accuracy on Kinetics-400 action recognition. With this work, we hope to bring greater attention to this promising path of freezing pretrained image models.
This paper describes how advanced deep learning based computer vision algorithms are applied to enable real-time on-board sensor processing for small UAVs. Four use cases are considered: target detection, classification and localization, road segmentation for autonomous navigation in GNSS-denied zones, human body segmentation, and human action recognition. All algorithms have been developed using state-of-the-art image processing methods based on deep neural networks. Acquisition campaigns have been carried out to collect custom datasets reflecting typical operational scenarios, where the peculiar point of view of a multi-rotor UAV is replicated. Algorithms architectures and trained models performances are reported, showing high levels of both accuracy and inference speed. Output examples and on-field videos are presented, demonstrating models operation when deployed on a GPU-powered commercial embedded device (NVIDIA Jetson Xavier) mounted on board of a custom quad-rotor, paving the way to enabling high level autonomy.
Young children are at an increased risk of contracting contagious diseases such as COVID-19 due to improper hand hygiene. An autonomous social agent that observes children while handwashing and encourages good hand washing practices could provide an opportunity for handwashing behavior to become a habit. In this article, we present a human action recognition system, which is part of the vision system of a social robot platform, to assist children in developing a correct handwashing technique. A modified convolution neural network (CNN) architecture with Channel Spatial Attention Bilinear Pooling (CSAB) frame, with a VGG-16 architecture as the backbone is trained and validated on an augmented dataset. The modified architecture generalizes well with an accuracy of 90% for the WHO-prescribed handwashing steps even in an unseen environment. Our findings indicate that the approach can recognize even subtle hand movements in the video and can be used for gesture detection and classification in social robotics.
Actions are about how we interact with the environment, including other people, objects, and ourselves. In this paper, we propose a novel multi-modal Holistic Interaction Transformer Network (HIT) that leverages the largely ignored, but critical hand and pose information essential to most human actions. The proposed "HIT" network is a comprehensive bi-modal framework that comprises an RGB stream and a pose stream. Each of them separately models person, object, and hand interactions. Within each sub-network, an Intra-Modality Aggregation module (IMA) is introduced that selectively merges individual interaction units. The resulting features from each modality are then glued using an Attentive Fusion Mechanism (AFM). Finally, we extract cues from the temporal context to better classify the occurring actions using cached memory. Our method significantly outperforms previous approaches on the J-HMDB, UCF101-24, and MultiSports datasets. We also achieve competitive results on AVA. The code will be available at https://github.com/joslefaure/HIT.
Traditional temporal action detection (TAD) usually handles untrimmed videos with small number of action instances from a single label (e.g., ActivityNet, THUMOS). However, this setting might be unrealistic as different classes of actions often co-occur in practice. In this paper, we focus on the task of multi-label temporal action detection that aims to localize all action instances from a multi-label untrimmed video. Multi-label TAD is more challenging as it requires for fine-grained class discrimination within a single video and precise localization of the co-occurring instances. To mitigate this issue, we extend the sparse query-based detection paradigm from the traditional TAD and propose the multi-label TAD framework of PointTAD. Specifically, our PointTAD introduces a small set of learnable query points to represent the important frames of each action instance. This point-based representation provides a flexible mechanism to localize the discriminative frames at boundaries and as well the important frames inside the action. Moreover, we perform the action decoding process with the Multi-level Interactive Module to capture both point-level and instance-level action semantics. Finally, our PointTAD employs an end-to-end trainable framework simply based on RGB input for easy deployment. We evaluate our proposed method on two popular benchmarks and introduce the new metric of detection-mAP for multi-label TAD. Our model outperforms all previous methods by a large margin under the detection-mAP metric, and also achieves promising results under the segmentation-mAP metric. Code is available at https://github.com/MCG-NJU/PointTAD.
In this paper, we study the transferability of ImageNet spatial and Kinetics spatio-temporal representations to multi-label Movie Trailer Genre Classification (MTGC). In particular, we present an extensive evaluation of the transferability of ConvNet and Transformer models pretrained on ImageNet and Kinetics to Trailers12k, a new manually-curated movie trailer dataset composed of 12,000 videos labeled with 10 different genres and associated metadata. We analyze different aspects that can influence transferability, such as frame rate, input video extension, and spatio-temporal modeling. In order to reduce the spatio-temporal structure gap between ImageNet/Kinetics and Trailers12k, we propose Dual Image and Video Transformer Architecture (DIViTA), which performs shot detection so as to segment the trailer into highly correlated clips, providing a more cohesive input for pretrained backbones and improving transferability (a 1.83% increase for ImageNet and 3.75% for Kinetics). Our results demonstrate that representations learned on either ImageNet or Kinetics are comparatively transferable to Trailers12k. Moreover, both datasets provide complementary information that can be combined to improve classification performance (a 2.91% gain compared to the top single pretraining). Interestingly, using lightweight ConvNets as pretrained backbones resulted in only a 3.46% drop in classification performance compared with the top Transformer while requiring only 11.82% of its parameters and 0.81% of its FLOPS.
In action recognition, although the combination of spatio-temporal videos and skeleton features can improve the recognition performance, a separate model and balancing feature representation for cross-modal data are required. To solve these problems, we propose Spatio-TemporAl cRoss (STAR)-transformer, which can effectively represent two cross-modal features as a recognizable vector. First, from the input video and skeleton sequence, video frames are output as global grid tokens and skeletons are output as joint map tokens, respectively. These tokens are then aggregated into multi-class tokens and input into STAR-transformer. The STAR-transformer encoder layer consists of a full self-attention (FAttn) module and a proposed zigzag spatio-temporal attention (ZAttn) module. Similarly, the continuous decoder consists of a FAttn module and a proposed binary spatio-temporal attention (BAttn) module. STAR-transformer learns an efficient multi-feature representation of the spatio-temporal features by properly arranging pairings of the FAttn, ZAttn, and BAttn modules. Experimental results on the Penn-Action, NTU RGB+D 60, and 120 datasets show that the proposed method achieves a promising improvement in performance in comparison to previous state-of-the-art methods.
Graph convolutional networks (GCNs), which can model the human body skeletons as spatial and temporal graphs, have shown remarkable potential in skeleton-based action recognition. However, in the existing GCN-based methods, graph-structured representation of the human skeleton makes it difficult to be fused with other modalities, especially in the early stages. This may limit their scalability and performance in action recognition tasks. In addition, the pose information, which naturally contains informative and discriminative clues for action recognition, is rarely explored together with skeleton data in existing methods. In this work, we propose pose-guided GCN (PG-GCN), a multi-modal framework for high-performance human action recognition. In particular, a multi-stream network is constructed to simultaneously explore the robust features from both the pose and skeleton data, while a dynamic attention module is designed for early-stage feature fusion. The core idea of this module is to utilize a trainable graph to aggregate features from the skeleton stream with that of the pose stream, which leads to a network with more robust feature representation ability. Extensive experiments show that the proposed PG-GCN can achieve state-of-the-art performance on the NTU RGB+D 60 and NTU RGB+D 120 datasets.
Temporal modeling is crucial for various video learning tasks. Most recent approaches employ either factorized (2D+1D) or joint (3D) spatial-temporal operations to extract temporal contexts from the input frames. While the former is more efficient in computation, the latter often obtains better performance. In this paper, we attribute this to a dilemma between the sufficiency and the efficiency of interactions among various positions in different frames. These interactions affect the extraction of task-relevant information shared among frames. To resolve this issue, we prove that frame-by-frame alignments have the potential to increase the mutual information between frame representations, thereby including more task-relevant information to boost effectiveness. Then we propose Alignment-guided Temporal Attention (ATA) to extend 1-dimensional temporal attention with parameter-free patch-level alignments between neighboring frames. It can act as a general plug-in for image backbones to conduct the action recognition task without any model-specific design. Extensive experiments on multiple benchmarks demonstrate the superiority and generality of our module.
Human action recognition in computer vision has been widely studied in recent years. However, most algorithms consider only certain action specially with even high computational cost. That is not suitable for practical applications with multiple actions to be identified with low computational cost. To meet various application scenarios, this paper presents a unified human action recognition framework composed of two modules, i.e., multi-form human detection and corresponding action classification. Among them, an open-source dataset is constructed to train a multi-form human detection model that distinguishes a human being's whole body, upper body or part body, and the followed action classification model is adopted to recognize such action as falling, sleeping or on-duty, etc. Some experimental results show that the unified framework is effective for various application scenarios. It is expected to be a new application-driven AI paradigm for human action recognition.
When applied to autonomous vehicle (AV) settings, action recognition can enhance an environment model's situational awareness. This is especially prevalent in scenarios where traditional geometric descriptions and heuristics in AVs are insufficient. However, action recognition has traditionally been studied for humans, and its limited adaptability to noisy, un-clipped, un-pampered, raw RGB data has limited its application in other fields. To push for the advancement and adoption of action recognition into AVs, this work proposes a novel two-stage action recognition system, termed RALACs. RALACs formulates the problem of action recognition for road scenes, and bridges the gap between it and the established field of human action recognition. This work shows how attention layers can be useful for encoding the relations across agents, and stresses how such a scheme can be class-agnostic. Furthermore, to address the dynamic nature of agents on the road, RALACs constructs a novel approach to adapting Region of Interest (ROI) Alignment to agent tracks for downstream action classification. Finally, our scheme also considers the problem of active agent detection, and utilizes a novel application of fusing optical flow maps to discern relevant agents in a road scene. We show that our proposed scheme can outperform the baseline on the ICCV2021 Road Challenge dataset and by deploying it on a real vehicle platform, we provide preliminary insight to the usefulness of action recognition in decision making.
This work is on training a generative action/video recognition model whose output is a free-form action-specific caption describing the video (rather than an action class label). A generative approach has practical advantages like producing more fine-grained and human-readable output, and being naturally open-world. To this end, we propose to adapt a pre-trained generative Vision & Language (V&L) Foundation Model for video/action recognition. While recently there have been a few attempts to adapt V&L models trained with contrastive learning (e.g. CLIP) for video/action, to the best of our knowledge, we propose the very first method that sets outs to accomplish this goal for a generative model. We firstly show that direct fine-tuning of a generative model to produce action classes suffers from severe overfitting. To alleviate this, we introduce REST, a training framework consisting of two key components: an unsupervised method for adapting the generative model to action/video by means of pseudo-caption generation and Self-training, i.e. without using any action-specific labels; (b) a Retrieval approach based on CLIP for discovering a diverse set of pseudo-captions for each video to train the model. Importantly, we show that both components are necessary to obtain high accuracy. We evaluate REST on the problem of zero-shot action recognition where we show that our approach is very competitive when compared to contrastive learning-based methods. Code will be made available.
We study the task of robust feature representations, aiming to generalize well on multiple datasets for action recognition. We build our method on Transformers for its efficacy. Although we have witnessed great progress for video action recognition in the past decade, it remains challenging yet valuable how to train a single model that can perform well across multiple datasets. Here, we propose a novel multi-dataset training paradigm, MultiTrain, with the design of two new loss terms, namely informative loss and projection loss, aiming to learn robust representations for action recognition. In particular, the informative loss maximizes the expressiveness of the feature embedding while the projection loss for each dataset mines the intrinsic relations between classes across datasets. We verify the effectiveness of our method on five challenging datasets, Kinetics-400, Kinetics-700, Moments-in-Time, Activitynet and Something-something-v2 datasets. Extensive experimental results show that our method can consistently improve state-of-the-art performance. Code and models are released.
Unmanned aerial vehicles (UAVs) are now widely applied to data acquisition due to its low cost and fast mobility. With the increasing volume of aerial videos, the demand for automatically parsing these videos is surging. To achieve this, current researches mainly focus on extracting a holistic feature with convolutions along both spatial and temporal dimensions. However, these methods are limited by small temporal receptive fields and cannot adequately capture long-term temporal dependencies which are important for describing complicated dynamics. In this paper, we propose a novel deep neural network, termed FuTH-Net, to model not only holistic features, but also temporal relations for aerial video classification. Furthermore, the holistic features are refined by the multi-scale temporal relations in a novel fusion module for yielding more discriminative video representations. More specially, FuTH-Net employs a two-pathway architecture: (1) a holistic representation pathway to learn a general feature of both frame appearances and shortterm temporal variations and (2) a temporal relation pathway to capture multi-scale temporal relations across arbitrary frames, providing long-term temporal dependencies. Afterwards, a novel fusion module is proposed to spatiotemporal integrate the two features learned from the two pathways. Our model is evaluated on two aerial video classification datasets, ERA and Drone-Action, and achieves the state-of-the-art results. This demonstrates its effectiveness and good generalization capacity across different recognition tasks (event classification and human action recognition). To facilitate further research, we release the code at https://gitlab.lrz.de/ai4eo/reasoning/futh-net.
Skeleton-based human action recognition has been drawing more interest recently due to its low sensitivity to appearance changes and the accessibility of more skeleton data. However, even the 3D skeletons captured in practice are still sensitive to the viewpoint and direction gave the occlusion of different human-body joints and the errors in human joint localization. Such view variance of skeleton data may significantly affect the performance of action recognition. To address this issue, we propose in this paper a new view-invariant representation learning approach, without any manual action labeling, for skeleton-based human action recognition. Specifically, we leverage the multi-view skeleton data simultaneously taken for the same person in the network training, by maximizing the mutual information between the representations extracted from different views, and then propose a global-local contrastive loss to model the multi-scale co-occurrence relationships in both spatial and temporal domains. Extensive experimental results show that the proposed method is robust to the view difference of the input skeleton data and significantly boosts the performance of unsupervised skeleton-based human action methods, resulting in new state-of-the-art accuracies on two challenging multi-view benchmarks of PKUMMD and NTU RGB+D.
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.
This paper presents OmniVL, a new foundation model to support both image-language and video-language tasks using one universal architecture. It adopts a unified transformer-based visual encoder for both image and video inputs, and thus can perform joint image-language and video-language pretraining. We demonstrate, for the first time, such a paradigm benefits both image and video tasks, as opposed to the conventional one-directional transfer (e.g., use image-language to help video-language). To this end, we propose a decoupled joint pretraining of image-language and video-language to effectively decompose the vision-language modeling into spatial and temporal dimensions and obtain performance boost on both image and video tasks. Moreover, we introduce a novel unified vision-language contrastive (UniVLC) loss to leverage image-text, video-text, image-label (e.g., image classification), video-label (e.g., video action recognition) data together, so that both supervised and noisily supervised pretraining data are utilized as much as possible. Without incurring extra task-specific adaptors, OmniVL can simultaneously support visual only tasks (e.g., image classification, video action recognition), cross-modal alignment tasks (e.g., image/video-text retrieval), and multi-modal understanding and generation tasks (e.g., image/video question answering, captioning). We evaluate OmniVL on a wide range of downstream tasks and achieve state-of-the-art or competitive results with similar model size and data scale.
Vision transformers are emerging as a powerful tool to solve computer vision problems. Recent techniques have also proven the efficacy of transformers beyond the image domain to solve numerous video-related tasks. Among those, human action recognition is receiving special attention from the research community due to its widespread applications. This article provides the first comprehensive survey of vision transformer techniques for action recognition. We analyze and summarize the existing and emerging literature in this direction while highlighting the popular trends in adapting transformers for action recognition. Due to their specialized application, we collectively refer to these methods as ``action transformers''. Our literature review provides suitable taxonomies for action transformers based on their architecture, modality, and intended objective. Within the context of action transformers, we explore the techniques to encode spatio-temporal data, dimensionality reduction, frame patch and spatio-temporal cube construction, and various representation methods. We also investigate the optimization of spatio-temporal attention in transformer layers to handle longer sequences, typically by reducing the number of tokens in a single attention operation. Moreover, we also investigate different network learning strategies, such as self-supervised and zero-shot learning, along with their associated losses for transformer-based action recognition. This survey also summarizes the progress towards gaining grounds on evaluation metric scores on important benchmarks with action transformers. Finally, it provides a discussion on the challenges, outlook, and future avenues for this research direction.