Conference paper Open Access

Unsupervised Video Summarization via Attention-Driven Adversarial Learning

Apostolidis, Evlampios; Adamantidou, Eleni; Metsai, Alexandros; Mezaris, Vasileios; Patras, Ioannis


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/56ad9cdf-0016-4488-9009-ee97a6bfd885/mmm2020_lncs11961_1_preprint.pdf"
      }, 
      "checksum": "md5:beaa35b2015dcdd1b1c3f0c86dcc9199", 
      "bucket": "56ad9cdf-0016-4488-9009-ee97a6bfd885", 
      "key": "mmm2020_lncs11961_1_preprint.pdf", 
      "type": "pdf", 
      "size": 978801
    }
  ], 
  "owners": [
    22750
  ], 
  "doi": "10.1007/978-3-030-37731-1_40", 
  "stats": {
    "version_unique_downloads": 171.0, 
    "unique_views": 515.0, 
    "views": 529.0, 
    "version_views": 529.0, 
    "unique_downloads": 171.0, 
    "version_unique_views": 515.0, 
    "volume": 178141782.0, 
    "version_downloads": 182.0, 
    "downloads": 182.0, 
    "version_volume": 178141782.0
  }, 
  "links": {
    "doi": "https://doi.org/10.1007/978-3-030-37731-1_40", 
    "latest_html": "https://zenodo.org/record/3605501", 
    "bucket": "https://zenodo.org/api/files/56ad9cdf-0016-4488-9009-ee97a6bfd885", 
    "badge": "https://zenodo.org/badge/doi/10.1007/978-3-030-37731-1_40.svg", 
    "html": "https://zenodo.org/record/3605501", 
    "latest": "https://zenodo.org/api/records/3605501"
  }, 
  "created": "2020-01-12T15:55:14.237496+00:00", 
  "updated": "2020-01-20T17:41:25.563084+00:00", 
  "conceptrecid": "3605500", 
  "revision": 3, 
  "id": 3605501, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.1007/978-3-030-37731-1_40", 
    "description": "<p>This paper presents a new video summarization approach that integrates an attention mechanism to identify the significant parts of the video, and is trained unsupervisingly via generative adversarial learning. Starting from the SUM-GAN model, we rst develop an improved version of it (called SUM-GAN-sl) that has a significantly reduced number of learned parameters, performs incremental training of the model&#39;s components, and applies a stepwise label-based strategy for updating the adversarial part. Subsequently, we introduce an attention mechanism to SUM-GAN-sl in two ways: i) by integrating an attention layer within the variational auto-encoder (VAE) of the architecture (SUM-GAN-VAAE), and ii) by replacing the VAE with a deterministic attention auto-encoder (SUM-GAN-AAE). Experimental evaluation on two datasets (SumMe and TVSum) documents the contribution of the attention auto-encoder to faster and more stable training of the model, resulting in a signicant performance improvement with respect to the original model and demonstrating the competitiveness of the proposed SUM-GAN-AAE against the state of the art.&nbsp;Software is publicly available at: https://github.com/e-apostolidis/SUM-GAN-AAE</p>", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "title": "Unsupervised Video Summarization via Attention-Driven Adversarial Learning", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "3605500"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "3605501"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "retv-h2020"
      }
    ], 
    "grants": [
      {
        "code": "780656", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100000780::780656"
        }, 
        "title": "Enhancing and Re-Purposing TV Content for Trans-Vector Engagement", 
        "acronym": "ReTV", 
        "program": "H2020", 
        "funder": {
          "doi": "10.13039/501100000780", 
          "acronyms": [], 
          "name": "European Commission", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100000780"
          }
        }
      }
    ], 
    "keywords": [
      "Video summarization", 
      "Unsupervised learning", 
      "Attention mechanism", 
      "Adversarial learning"
    ], 
    "publication_date": "2020-01-06", 
    "creators": [
      {
        "affiliation": "CERTH & QMUL", 
        "name": "Apostolidis, Evlampios"
      }, 
      {
        "affiliation": "CERTH", 
        "name": "Adamantidou, Eleni"
      }, 
      {
        "affiliation": "CERTH", 
        "name": "Metsai, Alexandros"
      }, 
      {
        "affiliation": "CERTH", 
        "name": "Mezaris, Vasileios"
      }, 
      {
        "affiliation": "QMUL", 
        "name": "Patras, Ioannis"
      }
    ], 
    "meeting": {
      "acronym": "MMM 2020", 
      "dates": "January 2020", 
      "place": "Daejeon, Korea", 
      "title": "26th Int. Conf. on Multimedia Modeling"
    }, 
    "access_right": "open", 
    "resource_type": {
      "subtype": "conferencepaper", 
      "type": "publication", 
      "title": "Conference paper"
    }
  }
}
529
182
views
downloads
Views 529
Downloads 182
Data volume 178.1 MB
Unique views 515
Unique downloads 171

Share

Cite as