AI Paper Atlas

Research map / Benchmarks / ScanObjectNN

ScanObjectNN leaderboard

25 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ScanObjectNN. Best accuracy: 97.2 (Ours, CVPR 2024).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

ScanObjectNN · accuracy

13 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
197.2OursOmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning
CVPR 2024 · Table 8
Cross-modal learning
290.63ReConContrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
ICML 2023 · Ablation study on pretraining
Self-supervised and contrastive
387.8MM-Point
DGCNN
MM-Point: Multi-View Information-Enhanced Multi-Modal Self-Supervised 3D Point Cloud Understanding
AAAI 2024 · Table: Evaluation of 3D point
Point cloud classification
487.2Ours w/ PN++ (Large)
PointNet++ (Large)
Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds
CVPR 2020 · Table: supervised methods on S
Point cloud classification
586.46ToThePoint
DGCNN
ToThePoint: Efficient Contrastive Learning of 3D Point Clouds via Recycling
CVPR 2023 · Table 2
Point cloud classification
681.7DGCNN + CrossPoint
DGCNN
CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding
CVPR 2022 · Table 2
Point cloud classification
766.32Duoduo CLIP (+MVA, 6 layers)
CLIP ViT-B/32
Duoduo CLIP: Efficient 3D Understanding with Multi-View Images
ICLR 2025 · Table: Multi-view attention ab
Shape and surface completion
864.67OmniBind-FullOmniBind: Large-scale Omni Multimodal Representation via Binding Spaces
ICLR 2025 · Zero-shot classification resul
Cross-modal learning
963.5OccTIP + DuoMamba
DuoMamba
Occlusion-aware Text-Image-Point Cloud Pretraining for Open-World 3D Object Recognition
CVPR 2025 · Table 1
Point cloud classification
1060.1VIT-LENS-G (OpenShape-Triplets)
OpenCLIP ViT-bigG/14
ViT-Lens: Towards Omni-modal Representations
CVPR 2024 · Table 2b
CLIP and zero-shot
3 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

ScanObjectNN · overall accuracy

12 papers, each with its best reported number for its own method; higher is better.

#overall accuracy ↑MethodPaperTopic
192.1CamPointCamPoint: Boosting Point Cloud Segmentation with Virtual Camera
CVPR 2025 · Table 8
Point cloud classification
290.7DeLA + DMR
DeLA
Dual Manifold Regularization Steered Robust Representation Learning for Point Cloud Analysis
AAAI 2025 · Table 4
Point cloud classification
390.7DeLA + X-3D
DeLA
X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition
CVPR 2024 · Table 3
Point cloud classification
489.2HGNet with self-attentionHGNet: Learning Hierarchical Geometry From Points, Edges, and Surfaces
CVPR 2023 · Table 5, row 7
Shape and surface completion
588.52ResRep + CP³
PointNeXt-S (C=64)
CP3: Channel Pruning Plug-In for Point-Based Networks
CVPR 2023 · Table 2
Point cloud classification
688.3PointNeXt+HyCoRe
PointNeXt
Rethinking the compositionality of point clouds through regularization in the hyperbolic space
NeurIPS 2022 · Table 2
Point cloud classification
788PCMPoint Cloud Mamba: Point Cloud Learning via State Space Model
AAAI 2025 · Table: Comparison with naive m
Point cloud classification
888PointNeXt + Ours
PointNeXt
Interpretable3D: An Ad-Hoc Interpretable Classifier for 3D Point Clouds
AAAI 2024 · Table 2
Point cloud classification
987.9Poly-PC (large)Poly-PC: A Polyhedral Network for Multiple Point Cloud Tasks at Once
CVPR 2023 · Table 9
Point cloud classification
1085.5SpikeNet-CSpikeNet: Sparse Spike-Driven Mask Vector Transformer for Energy-Efficient and Stable Spiking Point Cloud Processing
ICML 2026 · Table 1
Point cloud classification
2 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.