We propose a new method for learning videos by aggregating multiple models by sequentially extracting video clips from untrimmed video. The proposed method reduces the correlation between clips by feeding clips to multiple models in turn and synchronizes these models through federated learning. Experimental results show that the proposed method improves the performance compared to the no synchronization.
Action recognition is a well-established area of research in computer vision. In this paper, we propose S3Aug, a video data augmenatation for action recognition. Unlike conventional video data augmentation methods that involve cutting and pasting regions from two videos, the proposed method generates new videos from a single training video through segmentation and label-to-image transformation. Furthermore, the proposed method modifies certain categories of label images by sampling to generate a variety of videos, and shifts intermediate features to enhance the temporal coherency between frames of the generate videos. Experimental results on the UCF101, HMDB51, and Mimetics datasets demonstrate the effectiveness of the proposed method, paricularlly for out-of-context videos of the Mimetics dataset.
In this study, we propose a method for jointly learning of images and videos using a single model. In general, images and videos are often trained by separate models. We propose in this paper a method that takes a batch of images as input to Vision Transformer IV-ViT, and also a set of video frames with temporal aggregation by late fusion. Experimental results on two image datasets and two action recognition datasets are presented.
In this paper we propose an extension of the Attention Branch Network (ABN) by using instance segmentation for generating sharper attention maps for action recognition. Methods for visual explanation such as Grad-CAM usually generate blurry maps which are not intuitive for humans to understand, particularly in recognizing actions of people in videos. Our proposed method, Object-ABN, tackles this issue by introducing a new mask loss that makes the generated attention maps close to the instance segmentation result. Further the PC loss and multiple attention maps are introduced to enhance the sharpness of the maps and improve the performance of classification. Experimental results with UCF101 and SSv2 shows that the generated maps by the proposed method are much clearer qualitatively and quantitatively than those of the original ABN.
In the design of action recognition models, the quality of videos in the dataset is an important issue, however the trade-off between the quality and performance is often ignored. In general, action recognition models are trained and tested on high-quality videos, but in actual situations where action recognition models are deployed, sometimes it might not be assumed that the input videos are of high quality. In this study, we report qualitative evaluations of action recognition models for the quality degradation associated with transcoding by JPEG and H.264/AVC. Experimental results are shown for evaluating the performance of pre-trained models on the transcoded validation videos of Kinetics400. The models are also trained on the transcoded training videos. From these results, we quantitatively show the degree of degradation of the model performance with respect to the degradation of the video quality.
In this paper, we propose a multi-domain learning model for action recognition. The proposed method inserts domain-specific adapters between layers of domain-independent layers of a backbone network. Unlike a multi-head network that switches classification heads only, our model switches not only the heads, but also the adapters for facilitating to learn feature representations universal to multiple domains. Unlike prior works, the proposed method is model-agnostic and doesn't assume model structures unlike prior works. Experimental results on three popular action recognition datasets (HMDB51, UCF101, and Kinetics-400) demonstrate that the proposed method is more effective than a multi-head architecture and more efficient than separately training models for each domain.
We propose Multi-head Self/Cross-Attention (MSCA), which introduces a temporal cross-attention mechanism for action recognition, based on the structure of the Multi-head Self-Attention (MSA) mechanism of the Vision Transformer (ViT). Simply applying ViT to each frame of a video frame can capture frame features, but cannot model temporal features. However, simply modeling temporal information with CNN or Transfomer is computationally expensive. TSM that perform feature shifting assume a CNN and cannot take advantage of the ViT structure. The proposed model captures temporal information by shifting the Query, Key, and Value in the calculation of MSA of ViT. This is efficient without additional coinformationmputational effort and is a suitable structure for extending ViT over temporal. Experiments on Kineitcs400 show the effectiveness of the proposed method and its superiority over previous methods.
In this paper, we propose a data augmentation method for action recognition using instance segmentation. Although many data augmentation methods have been proposed for image recognition, few methods have been proposed for action recognition. Our proposed method, ObjectMix, extracts each object region from two videos using instance segmentation and combines them to create new videos. Experiments on two action recognition datasets, UCF101 and HMDB51, demonstrate the effectiveness of the proposed method and show its superiority over VideoMix, a prior work.
In this paper we propose a novel method which leverages the uncertainty measures provided by Bayesian deep networks through curriculum learning so that the uncertainty estimates are fed back to the system to resample the training data more densely in areas where uncertainty is high. We show in the concrete setting of a semantic segmentation task (iPS cell colony segmentation) that the proposed system is able to increase significantly the reliability of the model.
We propose a novel approach to identify the difficulty of visual questions for Visual Question Answering (VQA) without direct supervision or annotations to the difficulty. Prior works have considered the diversity of ground-truth answers of human annotators. In contrast, we analyze the difficulty of visual questions based on the behavior of multiple different VQA models. We propose to cluster the entropy values of the predicted answer distributions obtained by three different models: a baseline method that takes as input images and questions, and two variants that take as input images only and questions only. We use a simple k-means to cluster the visual questions of the VQA v2 validation set. Then we use state-of-the-art methods to determine the accuracy and the entropy of the answer distributions for each cluster. A benefit of the proposed method is that no annotation of the difficulty is required, because the accuracy of each cluster reflects the difficulty of visual questions that belong to it. Our approach can identify clusters of difficult visual questions that are not answered correctly by state-of-the-art methods. Detailed analysis on the VQA v2 dataset reveals that 1) all methods show poor performances on the most difficult cluster (about 10% accuracy), 2) as the cluster difficulty increases, the answers predicted by the different methods begin to differ, and 3) the values of cluster entropy are highly correlated with the cluster accuracy. We show that our approach has the advantage of being able to assess the difficulty of visual questions without ground-truth (i.e. the test set of VQA v2) by assigning them to one of the clusters. We expect that this can stimulate the development of novel directions of research and new algorithms. Clustering results are available online at https://github.com/tttamaki/vqd .