PulseAugur
LIVE 15:24:07
research · [2 sources] ·
0
research

New dataset and method improve video cropping for mobile viewing

Researchers have introduced a new dataset, LIVE-YouTube Video Cropping (LIVE-YT VC), to address the challenges of adapting landscape videos for mobile consumption. This database, featuring 1800 videos annotated by 90 human subjects, is the largest publicly available resource for subjective video cropping. The team also developed a smoothed version, LIVE-YT VC++, using a temporal filter to refine annotations. They demonstrated the dataset's utility with the SmartVidCrop algorithm and state-of-the-art video grounding models, aiming to establish it as a benchmark for future research. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Provides a new benchmark dataset for advancing video aspect ratio transformation models.

RANK_REASON Academic paper introducing a new dataset and methodology.

Read on arXiv cs.CV →

COVERAGE [2]

  1. arXiv cs.CV TIER_1 · Cheng-Han Lee, Maniratnam Mandal, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik ·

    Subjective Portrait Region Cropping in Landscape Videos with Temporal Annotation Smoothing

    arXiv:2604.24947v1 Announce Type: new Abstract: With the rise of mobile video consumption on diverse handheld display resolutions and orientation modes, altering videos to aspect ratios poses challenges. Static cropping and border padding often compromises visual quality, while w…

  2. arXiv cs.CV TIER_1 · Alan C. Bovik ·

    Subjective Portrait Region Cropping in Landscape Videos with Temporal Annotation Smoothing

    With the rise of mobile video consumption on diverse handheld display resolutions and orientation modes, altering videos to aspect ratios poses challenges. Static cropping and border padding often compromises visual quality, while warping may distort a video's intended meaning. H…