Pre-training for Action Recognition with Automatically Generated Fractal Datasets

• URL: http://arxiv.org/abs/2411.17584v1
• Date: Tue, 26 Nov 2024 16:51:11 GMT
• Title: Pre-training for Action Recognition with Automatically Generated Fractal Datasets
• Authors: Davyd Svyezhentsev, George Retsinas, Petros Maragos,
• Abstract summary: We present methods to automatically produce large-scale datasets of short synthetic video clips. The generated video clips are characterized by notable variety, stemmed by the innate ability of fractals to generate complex multi-scale structures. Compared to standard Kinetics pre-training, our reported results come close and are even superior on a portion of downstream datasets.
• Score: 23.686476742398973
• License:
• Abstract: In recent years, interest in synthetic data has grown, particularly in the context of pre-training the image modality to support a range of computer vision tasks, including object classification, medical imaging etc. Previous work has demonstrated that synthetic samples, automatically produced by various generative processes, can replace real counterparts and yield strong visual representations. This approach resolves issues associated with real data such as collection and labeling costs, copyright and privacy. We extend this trend to the video domain applying it to the task of action recognition. Employing fractal geometry, we present methods to automatically produce large-scale datasets of short synthetic video clips, which can be utilized for pre-training neural models. The generated video clips are characterized by notable variety, stemmed by the innate ability of fractals to generate complex multi-scale structures. To narrow the domain gap, we further identify key properties of real videos and carefully emulate them during pre-training. Through thorough ablations, we determine the attributes that strengthen downstream results and offer general guidelines for pre-training with synthetic videos. The proposed approach is evaluated by fine-tuning pre-trained models on established action recognition datasets HMDB51 and UCF101 as well as four other video benchmarks related to group action recognition, fine-grained action recognition and dynamic scenes. Compared to standard Kinetics pre-training, our reported results come close and are even superior on a portion of downstream datasets. Code and samples of synthetic videos are available at https://github.com/davidsvy/fractal_video .

Related papers

• Learning Video Representations without Natural Videos [36.0052738021796]
  We show that useful video representations can be learned from synthetic videos and natural images, without incorporating natural videos in the training. A VideoMAE model pre-trained on our synthetic videos closes 97.2% of the performance gap on UCF101 action classification between training from scratch and self-supervised pre-training from natural videos. Introducing crops of static images to the pre-training stage results in similar performance to UCF101 pre-training and outperforms the UCF101 pre-trained model on 11 out of 14 out-of-distribution datasets of UCF101-P.
  arXiv  Detail & Related papers  (2024-10-31T17:59:30Z)
• An Evaluation of Large Pre-Trained Models for Gesture Recognition using Synthetic Videos [32.257816070522885]
  We explore the possibility of using synthetically generated data for video-based gesture recognition with large pre-trained models. We use various state-of-the-art video encoders to extract features for use in k-nearest neighbors classification. We find that using synthetic training videos yields significantly lower classification accuracy on real test videos compared to using a relatively small number of real training videos.
  arXiv  Detail & Related papers  (2024-10-03T02:31:14Z)
• Appearance-Based Refinement for Object-Centric Motion Segmentation [85.2426540999329]
  We introduce an appearance-based refinement method that leverages temporal consistency in video streams to correct inaccurate flow-based proposals. Our approach involves a sequence-level selection mechanism that identifies accurate flow-predicted masks as exemplars. Its performance is evaluated on multiple video segmentation benchmarks, including DAVIS, YouTube, SegTrackv2, and FBMS-59.
  arXiv  Detail & Related papers  (2023-12-18T18:59:51Z)
• Learning Human Action Recognition Representations Without Real Humans [66.61527869763819]
  We present a benchmark that leverages real-world videos with humans removed and synthetic data containing virtual humans to pre-train a model. We then evaluate the transferability of the representation learned on this data to a diverse set of downstream action recognition benchmarks. Our approach outperforms previous baselines by up to 5%.
  arXiv  Detail & Related papers  (2023-11-10T18:38:14Z)
• CapST: An Enhanced and Lightweight Model Attribution Approach for Synthetic Videos [9.209808258321559]
  This paper investigates the model attribution problem of Deepfake videos from a recently proposed dataset, Deepfakes from Different Models (DFDM) The dataset comprises 6,450 Deepfake videos generated by five distinct models with variations in encoder, decoder, intermediate layer, input resolution, and compression ratio. Experimental results on the deepfake benchmark dataset (DFDM) demonstrate the efficacy of our proposed method, achieving up to a 4% improvement in accurately categorizing deepfake videos.
  arXiv  Detail & Related papers  (2023-11-07T08:05:09Z)
• Leaping Into Memories: Space-Time Deep Feature Synthesis [93.10032043225362]
  We propose LEAPS, an architecture-independent method for synthesizing videos from internal models. We quantitatively and qualitatively evaluate the applicability of LEAPS by inverting a range of architectures convolutional attention-based on Kinetics-400.
  arXiv  Detail & Related papers  (2023-03-17T12:55:22Z)
• Render In-between: Motion Guided Video Synthesis for Action Interpolation [53.43607872972194]
  We propose a motion-guided frame-upsampling framework that is capable of producing realistic human motion and appearance. A novel motion model is trained to inference the non-linear skeletal motion between frames by leveraging a large-scale motion-capture dataset. Our pipeline only requires low-frame-rate videos and unpaired human motion data but does not require high-frame-rate videos for training.
  arXiv  Detail & Related papers  (2021-11-01T15:32:51Z)
• HighlightMe: Detecting Highlights from Human-Centric Videos [52.84233165201391]
  We present a domain- and user-preference-agnostic approach to detect highlightable excerpts from human-centric videos. We use an autoencoder network equipped with spatial-temporal graph convolutions to detect human activities and interactions. We observe a 4-12% improvement in the mean average precision of matching the human-annotated highlights over state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-10-05T01:18:15Z)
• ViViT: A Video Vision Transformer [75.74690759089529]
  We present pure-transformer based models for video classification. Our model extracts-temporal tokens from the input video, which are then encoded by a series of transformer layers. We show how we can effectively regularise the model during training and leverage pretrained image models to be able to train on comparatively small datasets.
  arXiv  Detail & Related papers  (2021-03-29T15:27:17Z)
• Dynamic texture analysis for detecting fake faces in video sequences [6.1356022122903235]
  This work explores the analysis of texture-temporal dynamics of the video signal. The goal is to characterizing and distinguishing real fake sequences. We propose to build multiple binary decision on the joint analysis of temporal segments.
  arXiv  Detail & Related papers  (2020-07-30T07:21:24Z)
• Complex Human Action Recognition in Live Videos Using Hybrid FR-DL Method [1.027974860479791]
  We address challenges of the preprocessing phase, by an automated selection of representative frames among the input sequences. We propose a hybrid technique using background subtraction and HOG, followed by application of a deep neural network and skeletal modelling method. We name our model as Feature Reduction & Deep Learning based action recognition method, or FR-DL in short.
  arXiv  Detail & Related papers  (2020-07-06T15:12:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.