Logo InFlux and InFlux++

Leaderboard

Compare highlighted camera-intrinsics and LUT-Reliable EPE recall metrics. Select an evaluation set, then click the ↕, ▲, or ▼ arrow beside a metric threshold to choose the sort metric and direction. A submission not evaluated for the selected set remains visible with em dashes.

Higher recall is better ↑

Combined — Combined InFlux and InFlux++ Real evaluation sets
Sorted by LUT-Reliable EPE Recall @300 px, highest first. 8 public results.
Method fx Recall ↑ fy Recall ↑ cx Recall ↑ cy Recall ↑ LUT-Reliable EPE Recall ↑ Publication Code Date
@1% @10% @20% @1% @10% @20% @0.5% @1% @2% @0.5% @1% @2% @10 px @50 px @300 px
AnyCalib[2] 3.00 25.2 41.3 2.96 25.2 41.3 41.6 76.5 98.1 24.1 37.4 46.2 2.70 25.1 87.3 [paper] [code] 2026-07-25
AnyCalib (finetuned by InFlux++)[1] 3.68 31.2 50.4 3.65 31.3 50.5 33.0 67.2 95.4 20.4 33.4 44.1 1.93 18.7 78.6 [paper] [code] 2026-07-25
GeoCalib[5] 2.17 19.7 33.5 2.15 19.6 33.4 49.5 82.3 99.4 38.3 40.9 45.0 8.53e-1 16.7 75.4 [paper] [code] 2026-07-25
UniDepth v2[7] 1.41 14.0 26.6 1.31 13.1 25.7 21.8 45.4 69.9 3.87 9.14 25.5 5.68e-1 6.14 63.0 [paper] [code] 2026-07-25
WildCamera[8] 1.29 12.9 26.6 1.28 13.0 26.5 4.27 8.91 20.7 5.38 10.8 21.3 1.70e-1 4.26 59.1 [paper] [code] 2026-07-25
Perspective Fields[6] 7.32e-1 7.65 16.7 7.07e-1 7.53 16.5 1.48 2.97 5.93 1.55 3.09 6.20 1.29e-2 8.06e-1 23.7 [paper] [code] 2026-07-25
DroidCalib[4] 7.11e-1 5.99 11.8 5.66e-1 5.95 12.0 2.92 6.07 14.4 2.54 4.45 8.48 3.33e-1 1.86 16.3 [paper] [code] 2026-07-25
COLMAP[3] 1.49 6.34 7.22 1.56 6.36 7.23 3.14 7.16 8.87 1.78 1.96 2.31 2.25e-1 4.78 11.9 [paper] [code] 2026-07-25

Only results explicitly published by the original submitter are shown. Submitter email addresses and private prediction files are never exposed here.

References

Reference numbers are sorted by submission time.

  1. [1] InFlux++: Real and Synthetic Data for Estimating Dynamic Camera Intrinsics. [paper] [code]
  2. [2] AnyCalib: On-Manifold Learning for Model-Agnostic Single-View Camera Calibration. [paper] [code]
  3. [3] Structure-from-Motion Revisited. [paper] [code]
  4. [4] Deep Geometry-Aware Camera Self-Calibration from Video. [paper] [code]
  5. [5] GeoCalib: Learning Single-image Calibration with Geometric Optimization. [paper] [code]
  6. [6] Perspective Fields for Single Image Camera Calibration. [paper] [code]
  7. [7] UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler. [paper] [code]
  8. [8] Tame a Wild Camera: In-the-Wild Monocular Camera Calibration. [paper] [code]