In this survey, we present a systematic review of 3D hand pose estimation from the perspective of efficient annotation and learning. 3D hand pose estimation has been an important research area owing to its potential to enable various applications, such as video understanding, AR/VR, and robotics. However, the performance of models is tied to the quality and quantity of annotated 3D hand poses. Under the status quo, acquiring such annotated 3D hand poses is challenging, e.g., due to the difficulty of 3D annotation and the presence of occlusion. To reveal this problem, we review the pros and cons of existing annotation methods classified as manual, synthetic-model-based, hand-sensor-based, and computational approaches. Additionally, we examine methods for learning 3D hand poses when annotated data are scarce, including self-supervised pretraining, semi-supervised learning, and domain adaptation. Based on the study of efficient annotation and learning, we further discuss limitations and possible future directions in this field.
We introduce CVNets, a high-performance open-source library for training deep neural networks for visual recognition tasks, including classification, detection, and segmentation. CVNets supports image and video understanding tools, including data loading, data transformations, novel data sampling methods, and implementations of several standard networks with similar or better performance than previous studies. Our source code is available at: \url{https://github.com/apple/ml-cvnets}.
Video understanding has achieved great success in representation learning, such as video caption, video object grounding, and video descriptive question-answer. However, current methods still struggle on video reasoning, including evidence reasoning and commonsense reasoning. To facilitate deeper video understanding towards video reasoning, we present the task of Causal-VidQA, which includes four types of questions ranging from scene description (description) to evidence reasoning (explanation) and commonsense reasoning (prediction and counterfactual). For commonsense reasoning, we set up a two-step solution by answering the question and providing a proper reason. Through extensive experiments on existing VideoQA methods, we find that the state-of-the-art methods are strong in descriptions but weak in reasoning. We hope that Causal-VidQA can guide the research of video understanding from representation learning to deeper reasoning. The dataset and related resources are available at \url{https://github.com/bcmi/Causal-VidQA.git}.
Self-supervised learning has witnessed great progress in vision and NLP; recently, it also attracted much attention to various medical imaging modalities such as X-ray, CT, and MRI. Existing methods mostly focus on building new pretext self-supervision tasks such as reconstruction, orientation, and masking identification according to the properties of medical images. However, the publicly available self-supervision models are not fully exploited. In this paper, we present a powerful yet efficient self-supervision framework for surgical video understanding. Our key insight is to distill knowledge from publicly available models trained on large generic datasets4 to facilitate the self-supervised learning of surgical videos. To this end, we first introduce a semantic-preserving training scheme to obtain our teacher model, which not only contains semantics from the publicly available models, but also can produce accurate knowledge for surgical data. Besides training with only contrastive learning, we also introduce a distillation objective to transfer the rich learned information from the teacher model to self-supervised learning on surgical data. Extensive experiments on two surgical phase recognition benchmarks show that our framework can significantly improve the performance of existing self-supervised learning methods. Notably, our framework demonstrates a compelling advantage under a low-data regime. Our code is available at https://github.com/xmed-lab/DistillingSelf.
Temporal action detection (TAD) with end-to-end training often suffers from the pain of huge demand for computing resources due to long video duration. In this work, we propose an efficient temporal action detector (ETAD) that can train directly from video frames with extremely low GPU memory consumption. Our main idea is to minimize and balance the heavy computation among features and gradients in each training iteration. We propose to sequentially forward the snippet frame through the video encoder, and backward only a small necessary portion of gradients to update the encoder. To further alleviate the computational redundancy in training, we propose to dynamically sample only a small subset of proposals during training. Moreover, various sampling strategies and ratios are studied for both the encoder and detector. ETAD achieves state-of-the-art performance on TAD benchmarks with remarkable efficiency. On ActivityNet-1.3, training ETAD in 18 hours can reach 38.25% average mAP with only 1.3 GB memory consumption per video under end-to-end training. Our code will be publicly released.
Temporal action detection (TAD) is extensively studied in the video understanding community by generally following the object detection pipeline in images. However, complex designs are not uncommon in TAD, such as two-stream feature extraction, multi-stage training, complex temporal modeling, and global context fusion. In this paper, we do not aim to introduce any novel technique for TAD. Instead, we study a simple, straightforward, yet must-known baseline given the current status of complex design and low detection efficiency in TAD. In our simple baseline (termed BasicTAD), we decompose the TAD pipeline into several essential components: data sampling, backbone design, neck construction, and detection head. We extensively investigate the existing techniques in each component for this baseline, and more importantly, perform end-to-end training over the entire pipeline thanks to the simplicity of design. As a result, this simple BasicTAD yields an astounding and real-time RGB-Only baseline very close to the state-of-the-art methods with two-stream inputs. In addition, we further improve the BasicTAD by preserving more temporal and spatial information in network representation (termed as PlusTAD). Empirical results demonstrate that our PlusTAD is very efficient and significantly outperforms the previous methods on the datasets of THUMOS14 and FineAction. Meanwhile, we also perform in-depth visualization and error analysis on our proposed method and try to provide more insights on the TAD problem. Our approach can serve as a strong baseline for future TAD research. The code and model will be released at https://github.com/MCG-NJU/BasicTAD.
Human intelligence is multimodal; we integrate visual, linguistic, and acoustic signals to maintain a holistic worldview. Most current pretraining methods, however, are limited to one or two modalities. We present i-Code, a self-supervised pretraining framework where users may flexibly combine the modalities of vision, speech, and language into unified and general-purpose vector representations. In this framework, data from each modality are first given to pretrained single-modality encoders. The encoder outputs are then integrated with a multimodal fusion network, which uses novel attention mechanisms and other architectural innovations to effectively combine information from the different modalities. The entire system is pretrained end-to-end with new objectives including masked modality unit modeling and cross-modality contrastive learning. Unlike previous research using only video for pretraining, the i-Code framework can dynamically process single, dual, and triple-modality data during training and inference, flexibly projecting different combinations of modalities into a single representation space. Experimental results demonstrate how i-Code can outperform state-of-the-art techniques on five video understanding tasks and the GLUE NLP benchmark, improving by as much as 11% and demonstrating the power of integrative multimodal pretraining.
Building models that can be rapidly adapted to novel tasks using only a handful of annotated examples is an open challenge for multimodal machine learning research. We introduce Flamingo, a family of Visual Language Models (VLM) with this ability. We propose key architectural innovations to: (i) bridge powerful pretrained vision-only and language-only models, (ii) handle sequences of arbitrarily interleaved visual and textual data, and (iii) seamlessly ingest images or videos as inputs. Thanks to their flexibility, Flamingo models can be trained on large-scale multimodal web corpora containing arbitrarily interleaved text and images, which is key to endow them with in-context few-shot learning capabilities. We perform a thorough evaluation of our models, exploring and measuring their ability to rapidly adapt to a variety of image and video tasks. These include open-ended tasks such as visual question-answering, where the model is prompted with a question which it has to answer; captioning tasks, which evaluate the ability to describe a scene or an event; and close-ended tasks such as multiple-choice visual question-answering. For tasks lying anywhere on this spectrum, a single Flamingo model can achieve a new state of the art with few-shot learning, simply by prompting the model with task-specific examples. On numerous benchmarks, Flamingo outperforms models fine-tuned on thousands of times more task-specific data.
Long-form video understanding requires designing approaches that are able to temporally localize activities or language. End-to-end training for such tasks is limited by the compute device memory constraints and lack of temporal annotations at large-scale. These limitations can be addressed by pre-training on large datasets of temporally trimmed videos supervised by class annotations. Once the video encoder is pre-trained, it is common practice to freeze it during fine-tuning. Therefore, the video encoder does not learn temporal boundaries and unseen classes, causing a domain gap with respect to the downstream tasks. Moreover, using temporally trimmed videos prevents to capture the relations between different action categories and the background context in a video clip which results in limited generalization capacity. To address these limitations, we propose a novel post-pre-training approach without freezing the video encoder which leverages language. We introduce a masked contrastive learning loss to capture visio-linguistic relations between activities, background video clips and language in the form of captions. Our experiments show that the proposed approach improves the state-of-the-art on temporal action localization, few-shot temporal action localization, and video language grounding tasks.
Visual representation learning is ubiquitous in various real-world applications, including visual comprehension, video understanding, multi-modal analysis, human-computer interaction, and urban computing. Due to the emergence of huge amounts of multi-modal heterogeneous spatial/temporal/spatial-temporal data in big data era, the lack of interpretability, robustness, and out-of-distribution generalization are becoming the challenges of the existing visual models. The majority of the existing methods tend to fit the original data/variable distributions and ignore the essential causal relations behind the multi-modal knowledge, which lacks unified guidance and analysis about why modern visual representation learning methods easily collapse into data bias and have limited generalization and cognitive abilities. Inspired by the strong inference ability of human-level agents, recent years have therefore witnessed great effort in developing causal reasoning paradigms to realize robust representation and model learning with good cognitive ability. In this paper, we conduct a comprehensive review of existing causal reasoning methods for visual representation learning, covering fundamental theories, models, and datasets. The limitations of current methods and datasets are also discussed. Moreover, we propose some prospective challenges, opportunities, and future research directions for benchmarking causal reasoning algorithms in visual representation learning. This paper aims to provide a comprehensive overview of this emerging field, attract attention, encourage discussions, bring to the forefront the urgency of developing novel causal reasoning methods, publicly available benchmarks, and consensus-building standards for reliable visual representation learning and related real-world applications more efficiently.
For computer vision systems to operate in dynamic situations, they need to be able to represent and reason about object permanence. We introduce a framework for learning to estimate 4D visual representations from monocular RGB-D, which is able to persist objects, even once they become obstructed by occlusions. Unlike traditional video representations, we encode point clouds into a continuous representation, which permits the model to attend across the spatiotemporal context to resolve occlusions. On two large video datasets that we release along with this paper, our experiments show that the representation is able to successfully reveal occlusions for several tasks, without any architectural changes. Visualizations show that the attention mechanism automatically learns to follow occluded objects. Since our approach can be trained end-to-end and is easily adaptable, we believe it will be useful for handling occlusions in many video understanding tasks. Data, code, and models are available at https://occlusions.cs.columbia.edu/.
Spatio-temporal action detection is an important and challenging problem in video understanding. However, the application of the existing large-scale spatio-temporal action datasets in specific fields is limited, and there is currently no public tool for making spatio-temporal action datasets, it takes a lot of time and effort for researchers to customize the spatio-temporal action datasets, so we propose a multi-Person video dataset Annotation Method of spatio-temporally actions.First, we use ffmpeg to crop the videos and frame the videos; then use yolov5 to detect human in the video frame, and then use deep sort to detect the ID of the human in the video frame. By processing the detection results of yolov5 and deep sort, we can get the annotation file of the spatio-temporal action dataset to complete the work of customizing the spatio-temporal action dataset. https://github.com/Whiffe/Custom-ava-dataset_Custom-Spatio-Temporally-Action-Video-Dataset
The goal of this paper is to bypass the need for labelled examples in few-shot video understanding at run time. While proven effective, in many practical video settings even labelling a few examples appears unrealistic. This is especially true as the level of details in spatio-temporal video understanding and with it, the complexity of annotations continues to increase. Rather than performing few-shot learning with a human oracle to provide a few densely labelled support videos, we propose to automatically learn to find appropriate support videos given a query. We call this self-shot learning and we outline a simple self-supervised learning method to generate an embedding space well-suited for unsupervised retrieval of relevant samples. To showcase this novel setting, we tackle, for the first time, video instance segmentation in a self-shot (and few-shot) setting, where the goal is to segment instances at the pixel-level across the spatial and temporal domains. We provide strong baseline performances that utilize a novel transformer-based model and show that self-shot learning can even surpass few-shot and can be positively combined for further performance gains. Experiments on new benchmarks show that our approach achieves strong performance, is competitive to oracle support in some settings, scales to large unlabelled video collections, and can be combined in a semi-supervised setting.
Human action recognition (HAR) in videos is one of the core tasks of video understanding. Based on video sequences, the goal is to recognize actions performed by humans. While HAR has received much attention in the visible spectrum, action recognition in infrared videos is little studied. Accurate recognition of human actions in the infrared domain is a highly challenging task because of the redundant and indistinguishable texture features present in the sequence. Furthermore, in some cases, challenges arise from the irrelevant information induced by the presence of multiple active persons not contributing to the actual action of interest. Therefore, most existing methods consider a standard paradigm that does not take into account these challenges, which is in some part due to the ambiguous definition of the recognition task in some cases. In this paper, we propose a new method that simultaneously learns to recognize efficiently human actions in the infrared spectrum, while automatically identifying the key-actors performing the action without using any prior knowledge or explicit annotations. Our method is composed of three stages. In the first stage, optical flow-based key-actor identification is performed. Then for each key-actor, we estimate key-poses that will guide the frame selection process. A scale-invariant encoding process along with embedded pose filtering are performed in order to enhance the quality of action representations. Experimental results on InfAR dataset show that our proposed model achieves promising recognition performance and learns useful action representations.
Anomaly detection in videos is an important computer vision problem with various applications including automated video surveillance. Although adversarial attacks on image understanding models have been heavily investigated, there is not much work on adversarial machine learning targeting video understanding models and no previous work which focuses on video anomaly detection. To this end, we investigate an adversarial machine learning attack against video anomaly detection systems, that can be implemented via an easy-to-perform cyber-attack. Since surveillance cameras are usually connected to the server running the anomaly detection model through a wireless network, they are prone to cyber-attacks targeting the wireless connection. We demonstrate how Wi-Fi deauthentication attack, a notoriously easy-to-perform and effective denial-of-service (DoS) attack, can be utilized to generate adversarial data for video anomaly detection systems. Specifically, we apply several effects caused by the Wi-Fi deauthentication attack on video quality (e.g., slow down, freeze, fast forward, low resolution) to the popular benchmark datasets for video anomaly detection. Our experiments with several state-of-the-art anomaly detection models show that the attackers can significantly undermine the reliability of video anomaly detection systems by causing frequent false alarms and hiding physical anomalies from the surveillance system.
The objective of this paper is a temporal alignment network that ingests long term video sequences, and associated text sentences, in order to: (1) determine if a sentence is alignable with the video; and (2) if it is alignable, then determine its alignment. The challenge is to train such networks from large-scale datasets, such as HowTo100M, where the associated text sentences have significant noise, and are only weakly aligned when relevant. Apart from proposing the alignment network, we also make four contributions: (i) we describe a novel co-training method that enables to denoise and train on raw instructional videos without using manual annotation, despite the considerable noise; (ii) to benchmark the alignment performance, we manually curate a 10-hour subset of HowTo100M, totalling 80 videos, with sparse temporal descriptions. Our proposed model, trained on HowTo100M, outperforms strong baselines (CLIP, MIL-NCE) on this alignment dataset by a significant margin; (iii) we apply the trained model in the zero-shot settings to multiple downstream video understanding tasks and achieve state-of-the-art results, including text-video retrieval on YouCook2, and weakly supervised video action segmentation on Breakfast-Action; (iv) we use the automatically aligned HowTo100M annotations for end-to-end finetuning of the backbone model, and obtain improved performance on downstream action recognition tasks.
In this paper, we introduce a novel large-scale video dataset dubbed MM-SEAL for multi-person multi-grained spatio-temporal action localization among human daily life. We are the first to propose a new benchmark for multi-person spatio-temporal complex activity localization, where complex semantic and long duration bring new challenges to localization tasks. We observe that limited atomic actions can be combined into many complex activities. MM-SEAL provides both atomic action and complex activity annotations, producing 111.7k atomic actions spanning 172 action categories and 17.7k complex activities spanning 200 activity categories. We explore the relationship between atomic actions and complex activities, finding that atomic action features can improve the complex activity localization performance. Also, we propose a new network which generates temporal proposals and labels simultaneously, termed Faster-TAD. Finally, our evaluations show that visual features pretrained on MM-SEAL can improve the performance on other action localization benchmarks. We will release the dataset and the project code upon publication of the paper.
Temporal action detection (TAD) is an important yet challenging task in video understanding. It aims to simultaneously predict the semantic label and the temporal interval of every action instance in an untrimmed video. Rather than end-to-end learning, most existing methods adopt a head-only learning paradigm, where the video encoder is pre-trained for action classification, and only the detection head upon the encoder is optimized for TAD. The effect of end-to-end learning is not systematically evaluated. Besides, there lacks an in-depth study on the efficiency-accuracy trade-off in end-to-end TAD. In this paper, we present an empirical study of end-to-end temporal action detection. We validate the advantage of end-to-end learning over head-only learning and observe up to 11\% performance improvement. Besides, we study the effects of multiple design choices that affect the TAD performance and speed, including detection head, video encoder, and resolution of input videos. Based on the findings, we build a mid-resolution baseline detector, which achieves the state-of-the-art performance of end-to-end methods while running more than 4$\times$ faster. We hope that this paper can serve as a guide for end-to-end learning and inspire future research in this field. Code and models are available at \url{https://github.com/xlliu7/E2E-TAD}.
Most modern video recognition models are designed to operate on short video clips (e.g., 5-10s in length). Thus, it is challenging to apply such models to long movie understanding tasks, which typically require sophisticated long-range temporal reasoning. The recently introduced video transformers partially address this issue by using long-range temporal self-attention. However, due to the quadratic cost of self-attention, such models are often costly and impractical to use. Instead, we propose ViS4mer, an efficient long-range video model that combines the strengths of self-attention and the recently introduced structured state-space sequence (S4) layer. Our model uses a standard Transformer encoder for short-range spatiotemporal feature extraction, and a multi-scale temporal S4 decoder for subsequent long-range temporal reasoning. By progressively reducing the spatiotemporal feature resolution and channel dimension at each decoder layer, ViS4mer learns complex long-range spatiotemporal dependencies in a video. Furthermore, ViS4mer is $2.63\times$ faster and requires $8\times$ less GPU memory than the corresponding pure self-attention-based model. Additionally, ViS4mer achieves state-of-the-art results in $6$ out of $9$ long-form movie video classification tasks on the Long Video Understanding (LVU) benchmark. Furthermore, we show that our approach successfully generalizes to other domains, achieving competitive results on the Breakfast and the COIN procedural activity datasets. The code is publicly available at: https://github.com/md-mohaiminul/ViS4mer.
Visual private information leakage is an emerging key issue for the fast growing applications of video understanding like activity recognition. Existing approaches for mitigating privacy leakage in action recognition require privacy labels along with the action labels from the video dataset. However, annotating frames of video dataset for privacy labels is not feasible. Recent developments of self-supervised learning (SSL) have unleashed the untapped potential of the unlabeled data. For the first time, we present a novel training framework which removes privacy information from input video in a self-supervised manner without requiring privacy labels. Our training framework consists of three main components: anonymization function, self-supervised privacy removal branch, and action recognition branch. We train our framework using a minimax optimization strategy to minimize the action recognition cost function and maximize the privacy cost function through a contrastive self-supervised loss. Employing existing protocols of known-action and privacy attributes, our framework achieves a competitive action-privacy trade-off to the existing state-of-the-art supervised methods. In addition, we introduce a new protocol to evaluate the generalization of learned the anonymization function to novel-action and privacy attributes and show that our self-supervised framework outperforms existing supervised methods. Code available at: https://github.com/DAVEISHAN/SPAct