Proposes interpretable attention for video action recognition.
problem Efficient video action recognition with attention mechanisms.
method Spatial-temporal attention mechanism with saliency masks and convolutional LSTM.
result Improves video action recognition accuracy and spatial-temporal localization.
SVM with local features improves human action recognition.
problem Improving human action recognition in videos.
method Local appearance and motion features extracted using CNNs, concatenated, and used with SVM for classification.
result SVM with local features outperforms previous methods on benchmark datasets.
This research tackles few-shot video action recognition, improving accuracy with a two-stream setup.
problem Few-shot video action recognition with limited labeled examples.
method Two-stream models combining convolutional and recurrent neural network video encoders with metric-based few-shot algorithms.
result The setup achieves 84.2% accuracy on a 5-shot 5-way task, outperforming other methods.
Action recognition has attracted increasing attention from RGB input in computer vision partially due to potential applications on somatic simulation and statistics of sport such as virtual tennis game and tennis techniques and tactics analysis by video. Recently, deep learning based methods have achieved promising per…
Paper tackles zero-shot activity recognition using video features and text embeddings.
problem Zero-shot activity recognition with videos.
method Auto-encoder model for multimodal joint embedding, 3D convolutional action recognition for visual features, GloVe word embeddings for textual features.
result Improved zero-shot recognition results with top-n accuracy and mean Nearest Neighbor Overlap.
Improved action recognition in live videos with hybrid FR-DL method.
problem High computational costs and lack of temporal information in conventional action recognition.
method Automated selection of representative frames, feature extraction, background subtraction, HOG, deep neural network, LSTM, Softmax-KNN classifier.
result Significant improvement in accuracy and speed compared to state-of-the-art methods.
Paper presents a method for recognizing human actions using GLAC features from motion and static images.
problem Action recognition in 3D depth videos.
method 3D Motion Trail Model (3DMTM) for MHIs and SHIs, GLAC features extraction, l2-regularized Collaborative Representation Classifier (l2-CRC) for classification.
result The method outperforms other approaches in recognizing human actions.
Representation of human actions as a sequence of human body movements or action attributes enables the development of models for human activity recognition and summarization. We present an extension of the low-rank representation (LRR) model, termed the clustering-aware structure-constrained low-rank representation (CS…
With the rapid development of social media sharing, people often need to manage the growing volume of multimedia data such as large scale video classification and annotation, especially to organize those videos containing human activities. Recently, manifold regularized semi-supervised learning (SSL), which explores th…
Enhanced LSTM with multiple kernels and attention improves video action recognition.
problem Improving motion understanding in video analysis.
method Proposed a Network-in-LSTM approach with multiple convolutional kernels and layers, and an attention-based mechanism.
result Improves accuracy in supervised classification on UCF-101 and Sports-1M datasets.
A framework uses deep learning for activity recognition in IoT devices.
problem Activity recognition in IoT devices without physical contact.
method Background subtraction followed by 3D-Convolutional Neural Networks.
result Enhanced activity recognition using small IoT devices.
New dataset for UAV gesture recognition in outdoor settings.
problem Lack of public outdoor UAV gesture recognition datasets.
method Recorded 13 gestures in an outdoor setting, created 119 HD video clips, annotated with body joints and gesture classes.
result Baseline gesture recognition performance of 91.9% using P-CNN.
ESE-FN improves elderly activity recognition accuracy.
problem Recognizing individual actions and human-object interactions in elderly activities.
method Exploits multi-modal features from RGB videos and skeleton sequences using ESE attentions and a new Multi-modal Loss.
result ESE-FN achieves best accuracy on ETRI-Activity3D dataset.
Image understanding using deep convolutional network has reached human-level performance, yet a closely related problem of video understanding especially, action recognition has not reached the requisite level of maturity. We combine multi-kernels based support-vector-machines (SVM) with a multi-stream deep convolution…
Researchers create a flickering attack to fool video recognition networks.
problem Adversarial manipulation of video classification networks.
method Introducing a flickering temporal perturbation to fool video classifiers.
result Achieved high fooling ratio and temporal-invariant perturbation.
Limited annotated data available for the recognition of facial expression and action units embarrasses the training of deep networks, which can learn disentangled invariant features. However, a linear model with just several parameters normally is not demanding in terms of training data. In this paper, we propose an el…
PredNet fails to fully adhere to predictive coding principles.
problem Evaluation of PredNet's adherence to predictive coding theory.
method Critical analysis of PredNet's architecture and performance on a video action classification dataset.
result PredNet does not fully follow predictive coding principles but shows potential with top-down conditioning.
Improved speech recognition using EEG and video.
problem Enhancing continuous speech recognition systems.
method Implemented a CTC-based ASR model using EEG features.
result EEG features improve continuous visual speech recognition.
Recurrent Neural Networks (RNNs) are powerful sequence modeling tools. However, when dealing with high dimensional inputs, the training of RNNs becomes computational expensive due to the large number of model parameters. This hinders RNNs from solving many important computer vision tasks, such as Action Recognition in …
Generative model for high-resolution video generation.
problem Challenges in generating high-resolution videos due to memory and training stability limitations.
method Progressive growing of sliced Wasserstein GANs (SWGAN) for incremental spatiotemporal information learning.
result Generated photorealistic face videos of 256x256x32 resolution with an inception score of 14.57.
Improved action detection for multi-person videos using attention filtering.
problem Difficulty in distinguishing relevant parts of multi-person videos for action detection.
method Fovea attention filtering and generalized binary loss function.
result 20% relative improvement in mAP over baseline in AVA dataset.
Patient pain can be detected highly reliably from facial expressions using a set of facial muscle-based action units (AUs) defined by the Facial Action Coding System (FACS). A key characteristic of facial expression of pain is the simultaneous occurrence of pain-related AU combinations, whose automated detection would …
Recognizing group activities is challenging due to the difficulties in isolating individual entities, finding the respective roles played by the individuals and representing the complex interactions among the participants. Individual actions and group activities in videos can be represented in a common framework as the…
Recent studies have demonstrated the power of recurrent neural networks for machine translation, image captioning and speech recognition. For the task of capturing temporal structure in video, however, there still remain numerous open research questions. Current research suggests using a simple temporal feature pooling…
Morph accelerates 3D CNNs for video recognition, reducing energy consumption and improving performance.
problem Efficiently accelerating 3D CNNs for video recognition is challenging due to their large memory footprint and higher dimensionality.
method Designing a flexible accelerator called Morph that adapts to different spatial and temporal tiling strategies, and codesigning a software infrastructure to control the hardware.
result Morph achieves up to 3.4x reduction in energy consumption and up to 5.1x improvement in performance/watt compared to a baseline 3D CNN accelerator.
Proposes a new video attack method that multiplies perturbation to improve model robustness.
problem Challenges existing defense methods for video recognition models against additive adversarial attacks.
method Introduces Multiplicative Adversarial Videos (MultAV) to impose perturbation by multiplication.
result Model trained against additive attacks is less robust to MultAV.
V-BAD is a first black-box video attack framework that successfully fools deep video recognition models.
problem Vulnerability of video recognition models to black-box adversarial attacks.
method Tentative perturbations transferred from image models and partition-based rectifications for patches of tentative perturbations.
result V-BAD can craft both untargeted and targeted attacks with high success rates using a manageable number of queries.
Deep Fusion improves audio-video emotion recognition accuracy.
problem Challenges in automatic emotion recognition due to abstract concept and multiple expressions of emotion.
method Introduces factorized bilinear pooling (FBP) with embedded attention mechanism to integrate audio and video features.
result Achieves an accuracy of 62.48% on AFEW database, outperforming state-of-the-art results.
Recognising human activities from streaming videos poses unique challenges to learning algorithms: predictive models need to be scalable, incrementally trainable, and must remain bounded in size even when the data stream is arbitrarily long. Furthermore, as parameter tuning is problematic in a streaming setting, suitab…
Proposes a defense method against multiple adversarial video types.
problem Lack of multi-perturbation robustness in existing defense approaches.
method Adversarial training with multiple independent BN layers and a BN selection module.
result Demonstrates stronger multi-perturbation robustness against different adversarial video types.
Recent progress in using recurrent neural networks (RNNs) for image description has motivated the exploration of their application for video description. However, while images are static, working with videos requires modeling their dynamic temporal structure and then properly integrating that information into a natural…
Automated surgical workflow analysis and understanding can assist surgeons to standardize procedures and enhance post-surgical assessment and indexing, as well as, interventional monitoring. Computer-assisted interventional (CAI) systems based on video can perform workflow estimation through surgical instruments' recog…
Paper tackles activity recognition from body-worn video footage.
problem Classifying frames of body-worn video footage according to the wearer's activity.
method Extract motion features and semi-supervised classification.
result Method achieves comparable results to supervised and deep learning methods using less training data.
Paper investigates trade-off between video bitrate and CNN accuracy.
problem Network bandwidth and energy consumption issues in video classification.
method Only retain motion vector and select texture information at reduced bitrates.
result Achieved 11%-94% saving in bitrate with marginal accuracy loss.
This paper highlights the need for explainable AI in video deep learning models.
problem Lack of explainable AI methods for video deep learning models.
method Illustrates the current state of video deep learning and highlights the need for explainability methods.
result Current explainability methods for video deep learning are insufficient and need improvement.
A new method exposes motion-related relevance in video frames.
problem Deconstructing relevance in spatio-temporal models for video processing.
method Proposes a discriminative method to separate spatial and temporal relevance.
result Demonstrates effectiveness on UCF-101 action recognition dataset.
Paper explores activity recognition and prediction in real homes using sensor data and video.
problem Improving accuracy of activity recognition and prediction in real home environments.
method Binary sensor data, depth video data, field trial, probabilistic methods, LSTM networks, transfer learning, IIR filter.
result Achieved good accuracy in predicting next sensor event and its mean time of occurrence using LSTM model.
We propose a framework that learns a representation transferable across different domains and tasks in a label efficient manner. Our approach battles domain shift with a domain adversarial loss, and generalizes the embedding to novel task using a metric learning-based approach. Our model is simultaneously optimized on …
A fusion approach combines audio and video features for emotion recognition.
problem Continuous emotion recognition using both visual and auditory modalities.
method Pre-trained CNN features from video frames and minimalistic auditory descriptors. Fusion at feature or prediction level. SVR for prediction.
result Improves CCCs of 0.749 and 0.565 for arousal and valence respectively.
Paper proposes ICCN to learn correlations between text, audio, and video for multimodal sentiment analysis.
problem Improving multimodal sentiment analysis by learning hidden correlations between text and audio/video features.
method Interaction Canonical Correlation Network (ICCN) using deep canonical correlation analysis (DCCA).
result Empirical results confirm the effectiveness of ICCN in capturing useful information from all three views.
Paper proposes an active learning method for surgical workflow recognition using long-range temporal dependency.
problem Challenges in automatic surgical workflow recognition due to lack of large-scale labelled datasets.
method NL-RCNet with non-local block for capturing long-range temporal dependency and intra-clip dependency score for selection.
result Our approach outperforms state-of-the-art methods by selecting only 50% of samples for training.
New approach uses synthetic labels to train models on scarce annotated data for surgical phase recognition.
problem Learning surgical phase recognition from limited annotated data.
method Teacher/Student approach with a CNN-biLSTM-CRF teacher generating synthetic labels for a CNN-LSTM student.
result Improved surgical phase recognition performance with fewer annotated videos.
New method learns robot actions from videos without explicit labels.
problem Training robots to perform tasks from few demonstrations.
method Uses images and text for task-agnostic and general representation, synthesizes hallucinated actions, and applies dense correspondences.
result Trains robot policies solely from RGB videos, achieving diverse tasks across different robots and environments.
Current deep learning results on video generation are limited while there are only a few first results on video prediction and no relevant significant results on video completion. This is due to the severe ill-posedness inherent in these three problems. In this paper, we focus on human action videos, and propose a gene…
Pre-trains deep networks on random labels for improved supervised learning.
problem Improving supervised learning performance with limited labeled data.
method Randomly labels all samples in an unlabeled dataset and trains a deep neural network to memorize these labels.
result Improves classification accuracy by 1.5% to 5% on various action recognition datasets.
Simpler CNN model with spatial attention and temporal pooling outperforms complex models.
problem Emotion recognition from videos with small face deformations and identity variations.
method Spatial attention mechanism and temporal softmax pooling applied to a pre-trained CNN.
result The approach achieves higher accuracy than state-of-the-art methods on the EmotiW dataset.
Paper proposes a new LSTM model for spatio-temporal learning.
problem Challenging video tasks require learning long-term spatio-temporal correlations.
method Introduces a higher-order convolutional LSTM model with tensor train decomposition.
result Model achieves state-of-the-art performance with significantly fewer parameters.
We identify action representations from video data, proving their statistical benefits.
problem Identifying latent action policies from video data.
method Entropy-regularized LAPO objective, formalizing desiderata for action representations.
result Entropy-regularized LAPO identifies action representations satisfying desiderata under suitable conditions.