Condensing Action Segmentation Datasets via Generative Network Inversion

• URL: http://arxiv.org/abs/2503.14112v1
• Date: Tue, 18 Mar 2025 10:29:47 GMT
• Title: Condensing Action Segmentation Datasets via Generative Network Inversion
• Authors: Guodong Ding, Rongyu Chen, Angela Yao,
• Abstract summary: This work presents the first condensation approach for procedural video datasets used in temporal action segmentation.<n>We propose a condensation framework that leverages generative prior learned from the dataset and network inversion to condense data into compact latent codes.<n>Our evaluation on standard benchmarks demonstrates consistent effectiveness in condensing TAS datasets and achieving competitive performances.
• Score: 37.78120420622088
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This work presents the first condensation approach for procedural video datasets used in temporal action segmentation. We propose a condensation framework that leverages generative prior learned from the dataset and network inversion to condense data into compact latent codes with significant storage reduced across temporal and channel aspects. Orthogonally, we propose sampling diverse and representative action sequences to minimize video-wise redundancy. Our evaluation on standard benchmarks demonstrates consistent effectiveness in condensing TAS datasets and achieving competitive performances. Specifically, on the Breakfast dataset, our approach reduces storage by over 500$\times$ while retaining 83% of the performance compared to training with the full dataset. Furthermore, when applied to a downstream incremental learning task, it yields superior performance compared to the state-of-the-art.

Related papers

• PRISM: Video Dataset Condensation with Progressive Refinement and Insertion for Sparse Motion [22.804486552524885]
  This paper introduces PRISM, Progressive Refinement and Insertion for Sparse Motion, for video dataset condensation.<n>Unlike the previous method that separates static content from dynamic motion, our method preserves the essential interdependence between these elements.<n>Our approach progressively refines and inserts frames to fully accommodate the motion in an action while achieving better performance but less storage.
  arXiv  Detail & Related papers  (2025-05-28T16:42:10Z)
• Enhancing Generalization via Sharpness-Aware Trajectory Matching for Dataset Condensation [37.77454972709646]
  We introduce Sharpness-Aware Trajectory Matching (SATM), which enhances the generalization capability of learned synthetic datasets.<n>Our approach is mathematically well-supported and straightforward to implement along with controllable computational overhead.
  arXiv  Detail & Related papers  (2025-02-03T22:30:06Z)
• A Large-Scale Study on Video Action Dataset Condensation [35.194593167922804]
  We aim to bridge the gap between image and video dataset condensation by providing a large-scale study with systematic design and fair comparison.<n>Our work delves into three key aspects to provide valuable empirical insights: (1) temporal processing of video data, (2) the evaluation protocol for video dataset condensation, and (3) adaptation of condensation algorithms to the space-time domain.
  arXiv  Detail & Related papers  (2024-12-30T18:58:29Z)
• Video Set Distillation: Information Diversification and Temporal Densification [68.85010825225528]
  Video textbfsets have two dimensions of redundancies: within-sample and inter-sample redundancies. We are the first to study Video Set Distillation, which synthesizes optimized video data by addressing within-sample and inter-sample redundancies.
  arXiv  Detail & Related papers  (2024-11-28T05:37:54Z)
• Hierarchical Features Matter: A Deep Exploration of GAN Priors for Improved Dataset Distillation [51.44054828384487]
  We propose a novel parameterization method dubbed Hierarchical Generative Latent Distillation (H-GLaD) This method systematically explores hierarchical layers within the generative adversarial networks (GANs) In addition, we introduce a novel class-relevant feature distance metric to alleviate the computational burden associated with synthetic dataset evaluation.
  arXiv  Detail & Related papers  (2024-06-09T09:15:54Z)
• Coherent Temporal Synthesis for Incremental Action Segmentation [42.46228728930902]
  This paper presents the first exploration of video data replay techniques for incremental action segmentation. We propose a Temporally Coherent Action model, which represents actions using a generative model instead of storing individual frames. In a 10-task incremental setup on the Breakfast dataset, our approach achieves significant increases in accuracy for up to 22% compared to the baselines.
  arXiv  Detail & Related papers  (2024-03-10T06:07:06Z)
• Data Augmentation for Traffic Classification [54.92823760790628]
  Data Augmentation (DA) is a technique widely adopted in Computer Vision (CV) and Natural Language Processing (NLP) tasks. DA has struggled to gain traction in networking contexts, particularly in Traffic Classification (TC) tasks.
  arXiv  Detail & Related papers  (2024-01-19T15:25:09Z)
• Improved Distribution Matching for Dataset Condensation [91.55972945798531]
  We propose a novel dataset condensation method based on distribution matching. Our simple yet effective method outperforms most previous optimization-oriented methods with much fewer computational resources.
  arXiv  Detail & Related papers  (2023-07-19T04:07:33Z)
• DC-BENCH: Dataset Condensation Benchmark [79.18718490863908]
  This work provides the first large-scale standardized benchmark on dataset condensation. It consists of a suite of evaluations to comprehensively reflect the generability and effectiveness of condensation methods. The benchmark library is open-sourced to facilitate future research and application.
  arXiv  Detail & Related papers  (2022-07-20T03:54:05Z)
• Beyond Transfer Learning: Co-finetuning for Action Localisation [64.07196901012153]
  We propose co-finetuning -- simultaneously training a single model on multiple upstream'' and downstream'' tasks. We demonstrate that co-finetuning outperforms traditional transfer learning when using the same total amount of data. We also show how we can easily extend our approach to multiple upstream'' datasets to further improve performance.
  arXiv  Detail & Related papers  (2022-07-08T10:25:47Z)
• Dataset Condensation via Efficient Synthetic-Data Parameterization [40.56817483607132]
  Machine learning with massive amounts of data comes at a price of huge computation costs and storage for training and tuning. Recent studies on dataset condensation attempt to reduce the dependence on such massive data by synthesizing a compact training dataset. We propose a novel condensation framework that generates multiple synthetic data with a limited storage budget via efficient parameterization considering data regularity.
  arXiv  Detail & Related papers  (2022-05-30T09:55:31Z)
• Unsupervised Action Segmentation by Joint Representation Learning and Online Clustering [10.057155889852174]
  We present a novel approach for unsupervised activity segmentation which uses video frame clustering as a pretext task. We leverage temporal information in videos by employing temporal optimal transport. Our approach performs on par with or better than previous methods, despite having significantly less memory constraints.
  arXiv  Detail & Related papers  (2021-05-27T17:57:37Z)

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.