AI Paper Atlas

Research map / Point clouds / Point cloud classification

Symmetria: A Synthetic Dataset for Learning in Point Clouds

Ivan Sipiran, Gustavo Santelices, Lucas Oyarzún, Andrea Ranieri, Chiara Romanengo, Silvia Biasotti, B. Falcidieno

IJCV 2026 · Point clouds › Point cloud classification · approach: Self-supervised point cloud pretraining · 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
pre-train on SymSSL-10K instead of ShapeNetPoint-MAE · ModelNet40 k-NNaccuracy: 78.4 -> 81.7Improved
pre-train on SymSSL-10K instead of ShapeNetPoint-MAE · ModelNet40 fine-tuningaccuracy: 93.8 -> 92.5Hurt
pre-train on SymSSL-10K instead of ShapeNetPointGPT-S · ModelNet40 k-NNaccuracy: 81.6 -> 66.5Hurt
pre-train on SymSSL-10K instead of ShapeNetPointGPT-S · ScanObjectNN OBJ-ONLYaccuracy: 90.0 -> 79.0Hurt
pre-train on SymSSL-50K instead of ShapeNetPointGPT-S · ScanObjectNN PB-T50-RSaccuracy: 86.9 -> 84.9Mixed
pre-train on SymSSL-10K instead of ShapeNetPoint-MAE · ShapeNetPart instance mIoUmIoU: 86.1 -> 85.9No clear effect
increase SymSSL size from 10K to 50KPoint-MAE · ScanObjectNN OBJ-ONLYaccuracy: 87.8 -> 86.6Hurt
increase SymSSL size from 10K to 1MPoint-MAE · ModelNet40 k-NNaccuracy: 81.7 -> 73.3Hurt
increase SymSSL size from 10K to 1MPoint-MAE · ModelNet40 fine-tuningaccuracy: 92.5 -> 92.8No clear effect
add geometric perturbations (vs clean only) to SymSSL-10K (ablation)Point-MAE · ModelNet40 k-NNaccuracy: 78.6 -> 81.7Improved
add complex curve families (vs simple subset) to SymSSL-10K (ablation)Point-MAE · ModelNet40 k-NNaccuracy: 78.6 -> 79.6Improved
add geometric perturbations (vs clean only) to SymSSL-10K (ablation)Point-MAE · ShapeNetPart class mIoUmIoU: 83.7 -> 84.2Improved

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