Codesota · Computer Vision · Few-Shot Image Classification · NTIRE 2024 Transparent Surface Challenge (relative)Tasks/Computer Vision/Few-Shot Image Classification
Few-Shot Image Classification · benchmark dataset · EN

NTIRE 2024: HR Depth from Images of Specular and Transparent Surfaces (Booster Dataset) (Relative Depth).

The 'NTIRE 2024 Transparent Surface Challenge' dataset is part of the "NTIRE 2024: HR Depth from Images of Specular and Transparent Surfaces Challenge," held with the New Trends in Image Restoration and Enhancement (NTIRE) workshop at CVPR 2024. The challenge targets advancing depth estimation in challenging scenarios, specifically high-resolution images of specular and transparent (non-Lambertian) surfaces. The affiliated dataset, referred to as the 'Booster' dataset, provides annotated images focusing on these types of surfaces to foster algorithms capable of accurate, high-resolution depth prediction in difficult visual conditions. The dataset supports both monocular and stereo depth estimation tracks and is intended to catalyze the field toward solving unsolved challenges in depth prediction.

Paper Submit a result
§ 01 · Leaderboard

Best published scores.

No results indexed yet — be the first to submit a score.

No benchmark results indexed yet
§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies