AI Paper Atlas

Research map / 3D vision / 3D generation

ReassembleNet: Learnable Keypoints and Diffusion for 2D Fresco Reconstruction

Adeela Islam, Stefano Fiorini, Stuart James, Pietro Morerio, Alessio Del Bue

ICCV 2025 · 3D vision › 3D generation · approach: Feed-forward 3D asset generation and editing · 6 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 learnable keypoint selector instead of farthest point sampling (ablation)ReassembleNet (fine-tuned) · RePAIR 2DRMSE rotation (deg): 42.98 -> 32.91Improved
add learnable keypoint selector instead of farthest point sampling (ablation)ReassembleNet (no fine-tuning) · RePAIR 2DRMSE rotation (deg): 55.01 -> 47.61 rotation RMSE; Qpos 0.35 -> 0.27Mixed
add trainable instead of frozen keypoint selector (ablation)ReassembleNet (fine-tuned) · RePAIR 2DRMSE translation (mm): 18.41 -> 17.18Improved
add fine-tuning on RePAIR after semi-synthetic pretraining (ablation)ReassembleNet w/ Learnable KP selection · RePAIR 2DRMSE rotation (deg): 47.61 -> 32.91Improved
add fine-tuning after pretraining (ablation)ReassembleNet w/ Learnable KP selection · RePAIR 2DQpos: 0.27 -> 0.21Hurt
increase number of keypoints k from 3 to 20 (ablation)ReassembleNet · RePAIR 2DRMSE rotation (deg): 103.26 -> 47.61Improved
increase number of keypoints k from 3 to 20 (ablation)ReassembleNet · RePAIR 2DGPU memory (MB): 22630 -> 36701Hurt
use erosion and random jitter in semi-synthetic fragment generation (ablation)ReassembleNet pretraining · RePAIR 2DRMSE rotation (deg): 56.54 -> 32.91Improved
use erosion and random jitter in semi-synthetic fragment generation (ablation)ReassembleNet pretraining · RePAIR 2DRMSE translation (mm): 26.42 -> 17.18Improved
add geometric, local and global texture features to learnable-selector model (no transfer) (ablation)ReassembleNet Frozen KP selection · RePAIR 2DRMSE rotation (deg): 62.45 -> 51.96Improved
add all features and transfer learning (ablation)ReassembleNet w/ Learnable KP selection · RePAIR 2DRMSE rotation (deg): 47.61 -> 32.91Improved
treat pieces as keypoint sets instead of square patchesReassembleNet vs DiffAssemble · RePAIR 2D (fine-tuning)RMSE rotation (deg): 123.42 -> 32.91Improved

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