AI Paper Atlas

Research map / 3D vision / Gaussian splatting

ThermoGS: Decoupling Physical Surface Attributes for Spatio-Temporal Thermal Field Emulation via 4D Gaussian Splatting

Kun Yang, Yuxiang Liu, Zeyu Cui, Shen Yan, Maojun Zhang, Yu Liu, Xue Wang, Qing Wang

ECCV 2026 · 3D vision › Gaussian splatting · approach: Sparse-view Gaussian splatting for novel views · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add all five loss terms (edge, smooth, physical, ODE, PINN) instead of dropping any single one (ablation)ThermoGS · Thermo S1-S4temperature MAE (C): 1.147 -> 0.932Improved
add all five loss terms instead of dropping any single one (ablation)ThermoGS · Thermo S1-S4PSNR (dB): 31.17 -> 31.83Improved
predict physical attributes with PINN instead of table lookup (ablation)ThermoGS · Thermo S1-S4temperature MAE (C): 1.282 -> 0.932Improved
predict physical attributes with PINN instead of table lookup (ablation)ThermoGS · Thermo S1-S4PSNR (dB): 31.06 -> 31.83Improved
ThermoGS vs NTR-Gaussian on full-day cross-time/scene testThermoGS · Thermo S1-S4temperature MAE (C): 1.136 -> 0.932Improved
ThermoGS vs NTR-Gaussian on S3 held-out scenesThermoGS · Thermo S3 (S1,S2,S4)temperature MAE (C): 3.122 -> 1.49Improved
ThermoGS vs NTR-Gaussian on S4 held-out scenesThermoGS · Thermo S4 (S1,S2,S3)PSNR (dB): 28.80 -> 28.81No clear effect
ThermoGS vs 4DGS on S3 held-out scenesThermoGS · Thermo S3 (S1,S2,S4)SSIM: 0.949 -> 0.955No clear effect
ThermoGS vs NTR-Gaussian on nighttime NTR dataset S2ThermoGS · NTR Dataset nighttime S2temperature MAE (C): 1.709 -> 0.785Improved
ThermoGS vs NTR-Gaussian on nighttime NTR dataset S4ThermoGS · NTR Dataset nighttime S4temperature MAE (C): 1.065 -> 0.896Improved
infer attributes with PINN rather than NTR-Gaussian approach: emissivityThermoGS · Thermo physical attribute inferenceMAE of emissivity vs reference: 0.470 -> 0.018Improved
infer attributes: convective coefficient hThermoGS · Thermo physical attribute inferenceMAE of h vs reference: 32.80 -> 0.534Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy