Codesota · Computer Vision · Few-Shot Image Classification · Tanks and Temples (6)Tasks/Computer Vision/Few-Shot Image Classification
Few-Shot Image Classification · benchmark dataset · EN

Tanks and Temples: Benchmarking Large-Scale Scene Reconstruction.

Tanks and Temples is a widely-used benchmark for image-based 3D reconstruction and multi-view stereo (MVS). Introduced by Knapitsch et al. (Tanks and Temples: Benchmarking Large-Scale Scene Reconstruction), the benchmark provides high-resolution video/image sequences of real-world scenes and laser-scanned ground-truth geometry for evaluating reconstruction and novel-view-synthesis methods. The benchmark is organized into testing groups (commonly referred to as the “intermediate” and “advanced” sets) and is frequently used in the literature; many papers also evaluate on a standard 6-scene subset of the benchmark for out-of-domain novel-view-synthesis (NVS) evaluation. Official dataset/download and benchmark materials are hosted at the TanksAndTemples website and the original paper (SIGGRAPH / ACM TOG) provides dataset details and evaluation instructions.

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