AI Paper Atlas

Research map / Point clouds / Registration and matching

Diff2I2P: Differentiable Image-to-Point Cloud Registration with Diffusion Prior

Juncheng Mu, Chengwei Ren, Weixiang Zhang, Liang Pan, Xiao-Ping Zhang, Yue Gao

ICCV 2025 · Point clouds › Registration and matching · approach: Learned feature-based registration

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 CSD loss (diffusion score distillation) (ablation)Diff2I2P · 7-ScenesRegistration Recall (%): 81.3 -> 83.0Improved
use overlap mask in CSD loss (ablation)Diff2I2P · 7-ScenesRegistration Recall (%): 81.7 -> 83.0Improved
add augmented circle loss instead of plain circle loss (ablation)Diff2I2P · 7-ScenesRegistration Recall (%): 82.5 -> 83.0Improved
add transformation loss (MSE of projected points) (ablation)Diff2I2P · 7-Scenestraining stability: Gradient ExplosionHurt
use Diff2I2P instead of 2D3D-MATR2D3D-MATR · 7-ScenesRegistration Recall (%): 75.8 -> 83.0Improved
distill diffusion features into backbone instead of using diffusion features at inference (vs FreeReg)FreeReg · 7-ScenesRegistration Recall (%) / time: 71.2 -> 83.0; 9.634 s -> 0.074 sImproved
perturb input point clouds (shifts, 1% randomized, 10% removed)Diff2I2P · 7-ScenesRegistration Recall (%): 83.0 -> 82.5 / 82.0 / 81.8No clear effect
apply Diff2I2P to outdoor KITTI instead of 2D3D-MATR2D3D-MATR · KITTIRegistration Recall (%): 75.4 -> 82.2Improved
Diff2I2P vs 2D3D-MATR on feature matching recall2D3D-MATR · 7-ScenesFeature Matching Recall (%): 92.1 -> 92.2No clear effect

Later papers that compared with this method

2 later accepted papers report Diff2I2P as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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