{
  "video": "video-09dc4775.mp4",
  "description": "This video is a presentation, likely a conference talk or a paper demonstration, titled **\"See-through: Single-Image Layer Decomposition for Anime Characters\"**.\n\nHere is a detailed breakdown of what is happening:\n\n**Visual Content:**\nThe primary visual element is a slide presentation displayed on the screen.\n\n*   **Title Slide:** The title, **\"See-through: Single-Image Layer Decomposition for Anime Characters\"**, is prominent.\n*   **Authors and Affiliations:** A list of authors is visible, including **Ian Liu, Chengzhe Li, Haoyuan Qiu, Kun Wang Chan, Yanghua Jia, Hanyuan Liu, Stephen Chun Wang,** and **Xuejing Liu**, along with their affiliations (e.g., Saint Francis University, University of Pennsylvania, Spellbrush, Shitagaki Lab).\n*   **Context:** The slide notes that the work is \"Conditionally accepted to appear in ACM SIGGRAPH 2026 Conference Proceedings.\"\n*   **Abstract/Summary (TL;DR):** A brief summary, titled **\"TL;DR\"**, describes the research: \"We introduce a framework that automates the transformation of static anime illustrations into manipulable 2.5D models. Our approach decomposes a single image into fully inpainted, semantically distinct layers with inferred drawing orders\u2014up to 23 layers including hair, face, eyes, clothing, accessories, and more.\"\n*   **Visual Examples:** Below the text, there is a row of image thumbnails showing various anime-style character renderings, suggesting the input and output of the system being presented.\n*   **Code/Repository Information:** In the upper left corner, there are badges indicating the repository (**\"arxiv: 2602.03749\"**) and options to view it on **\"PDF\"** or **\"Insert Here\"**.\n\n**Audio/Narration (Implied):**\nSince this is a video recording, there is likely accompanying narration explaining the research, detailing the problem (static images vs. manipulable 2.5D models), the proposed solution (layer decomposition), and the results shown in the images.\n\n**Overall Purpose:**\nThe video serves as a technical presentation introducing and demonstrating a research contribution in computer graphics or artificial intelligence, specifically focusing on using a single 2D image of an anime character to automatically separate it into many distinct, editable layers, thereby enabling it to be treated as a more complex, manipulable 2.5D asset.",
  "codec": "av1",
  "transcoded": true,
  "elapsed_s": 12.7
}