{"url":"/sota/3d-point-cloud-classification-on-scanobjectnn","task":{"name":"3D Point Cloud Classification","url":"/task/3d-point-cloud-classification","note":null},"dataset":{"name":"ScanObjectNN","url":"/dataset/scanobjectnn"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":null,"description_from":null,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Overall Accuracy","Mean Accuracy","OBJ-BG (OA)","OBJ-ONLY (OA)","FLOPs","Number of params"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Overall Accuracy":"higher","Mean Accuracy":"higher","OBJ-BG (OA)":null,"OBJ-ONLY (OA)":null,"FLOPs":"lower","Number of params":"lower"}},"counts":{"rows":77,"rows_with_code":72,"rows_with_paper_page":76,"rows_dated":77,"rows_using_additional_data":31},"rows":[{"rank_in_archive_order":1,"model":"OmniVec2","metrics":{"Overall Accuracy":"97.2"},"uses_additional_data":true,"paper_date":"2024-01-01","paper":"/paper/omnivec2-a-novel-transformer-based-network","paper_url":"http://openaccess.thecvf.com//content/CVPR2024/html/Srivastava_OmniVec2_-_A_Novel_Transformer_based_Network_for_Large_Scale_CVPR_2024_paper.html","paper_title":"OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"PointGST","metrics":{"OBJ-BG (OA)":"99.48","OBJ-ONLY (OA)":"97.76","Overall Accuracy":"96.18"},"uses_additional_data":true,"paper_date":"2024-10-10","paper":"/paper/parameter-efficient-fine-tuning-in-spectral","paper_url":"https://arxiv.org/abs/2410.08114v1","paper_title":"Parameter-Efficient Fine-Tuning in Spectral Domain for Point Cloud Learning","code":"https://github.com/jerryfeng2003/pointgst","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":1,"n_samples":13,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"OmniVec","metrics":{"Overall Accuracy":"96.1"},"uses_additional_data":true,"paper_date":"2023-11-07","paper":"/paper/omnivec-learning-robust-representations-with","paper_url":"https://arxiv.org/abs/2311.05709v1","paper_title":"OmniVec: Learning robust representations with cross modal sharing","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"GPSFormer","metrics":{"FLOPs":"0.7G","Mean Accuracy":"93.8","Number of params":"2.36M","Overall Accuracy":"95.4"},"uses_additional_data":false,"paper_date":"2024-07-18","paper":"/paper/gpsformer-a-global-perception-and-local","paper_url":"https://arxiv.org/abs/2407.13519v2","paper_title":"GPSFormer: A Global Perception and Local Structure Fitting-based Transformer for Point Cloud Understanding","code":"https://github.com/changshuowang/GPSFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"ReCon++","metrics":{"OBJ-BG (OA)":"98.80","OBJ-ONLY (OA)":"97.59","Overall Accuracy":"95.25"},"uses_additional_data":true,"paper_date":"2024-02-27","paper":"/paper/shapellm-universal-3d-object-understanding","paper_url":"https://arxiv.org/abs/2402.17766v3","paper_title":"ShapeLLM: Universal 3D Object Understanding for Embodied Interaction","code":"https://github.com/qizekun/ShapeLLM","n_code_links":3,"syntology":{"n_ran":9,"n_unverified":8,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"AsymDSD-B* (no voting)","metrics":{"OBJ-BG (OA)":"96.73","OBJ-ONLY (OA)":"94.32","Overall Accuracy":"93.72"},"uses_additional_data":true,"paper_date":"2025-06-26","paper":"/paper/asymmetric-dual-self-distillation-for-3d-self","paper_url":"https://arxiv.org/abs/2506.21724v1","paper_title":"Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning","code":"https://github.com/RFLeijenaar/AsymDSD","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"PointGPT","metrics":{"OBJ-BG (OA)":"97.2","OBJ-ONLY (OA)":"96.6","Overall Accuracy":"93.4"},"uses_additional_data":true,"paper_date":"2023-05-19","paper":"/paper/pointgpt-auto-regressively-generative-pre-1","paper_url":"https://arxiv.org/abs/2305.11487v2","paper_title":"PointGPT: Auto-regressively Generative Pre-training from Point Clouds","code":"https://github.com/CGuangyan-BIT/PointGPT","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"GPSFormer-elite","metrics":{"Mean Accuracy":"92.51","Number of params":"0.68M","Overall Accuracy":"93.30"},"uses_additional_data":false,"paper_date":"2024-07-18","paper":"/paper/gpsformer-a-global-perception-and-local","paper_url":"https://arxiv.org/abs/2407.13519v2","paper_title":"GPSFormer: A Global Perception and Local Structure Fitting-based Transformer for Point Cloud Understanding","code":"https://github.com/changshuowang/GPSFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"Mamba3D","metrics":{"FLOPs":"3.9G","Number of params":"16.9M","OBJ-BG (OA)":"94.49","OBJ-ONLY (OA)":"92.43","Overall Accuracy":"92.64"},"uses_additional_data":false,"paper_date":"2024-04-23","paper":"/paper/mamba3d-enhancing-local-features-for-3d-point","paper_url":"https://arxiv.org/abs/2404.14966v2","paper_title":"Mamba3D: Enhancing Local Features for 3D Point Cloud Analysis via State Space Model","code":"https://github.com/xhanxu/Mamba3D","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":10,"model":"Mamba3D (no voting)","metrics":{"FLOPs":"3.9G","Number of params":"16.9M","OBJ-BG (OA)":"92.94","OBJ-ONLY (OA)":"92.08","Overall Accuracy":"91.81"},"uses_additional_data":false,"paper_date":"2024-04-23","paper":"/paper/mamba3d-enhancing-local-features-for-3d-point","paper_url":"https://arxiv.org/abs/2404.14966v2","paper_title":"Mamba3D: Enhancing Local Features for 3D Point Cloud Analysis via State Space Model","code":"https://github.com/xhanxu/Mamba3D","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":11,"model":"ULIP-2 + PointNeXt","metrics":{"Mean Accuracy":"91.2","Number of params":"1.4M","Overall Accuracy":"91.5"},"uses_additional_data":true,"paper_date":"2023-05-14","paper":"/paper/ulip-2-towards-scalable-multimodal-pre","paper_url":"https://arxiv.org/abs/2305.08275v4","paper_title":"ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding","code":"https://github.com/salesforce/ulip","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":12,"model":"ReCon","metrics":{"OBJ-BG (OA)":"95.35","OBJ-ONLY (OA)":"93.80","Overall Accuracy":"91.26"},"uses_additional_data":true,"paper_date":"2023-02-05","paper":"/paper/contrast-with-reconstruct-contrastive-3d","paper_url":"https://arxiv.org/abs/2302.02318v2","paper_title":"Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining","code":"https://github.com/qizekun/ReCon","n_code_links":5,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"ULIP-2 + PointNeXt (no voting)","metrics":{"Mean Accuracy":"90.3","Number of params":"1.4M","Overall Accuracy":"90.8"},"uses_additional_data":true,"paper_date":"2023-05-14","paper":"/paper/ulip-2-towards-scalable-multimodal-pre","paper_url":"https://arxiv.org/abs/2305.08275v4","paper_title":"ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding","code":"https://github.com/salesforce/ulip","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":14,"model":"ReCon (no voting)","metrics":{"OBJ-BG (OA)":"95.18","OBJ-ONLY (OA)":"93.29","Overall Accuracy":"90.63"},"uses_additional_data":true,"paper_date":"2023-02-05","paper":"/paper/contrast-with-reconstruct-contrastive-3d","paper_url":"https://arxiv.org/abs/2302.02318v2","paper_title":"Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining","code":"https://github.com/qizekun/ReCon","n_code_links":5,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"AsymDSD-S (no voting)","metrics":{"OBJ-BG (OA)":"94.32","OBJ-ONLY (OA)":"91.91","Overall Accuracy":"90.53"},"uses_additional_data":true,"paper_date":"2025-06-26","paper":"/paper/asymmetric-dual-self-distillation-for-3d-self","paper_url":"https://arxiv.org/abs/2506.21724v1","paper_title":"Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning","code":"https://github.com/RFLeijenaar/AsymDSD","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"DeLA","metrics":{"FLOPs":"1.5G","Mean Accuracy":"89.3","Number of params":"5.3M","Overall Accuracy":"90.4"},"uses_additional_data":false,"paper_date":"2023-08-31","paper":"/paper/decoupled-local-aggregation-for-point-cloud","paper_url":"https://arxiv.org/abs/2308.16532v1","paper_title":"Decoupled Local Aggregation for Point Cloud Learning","code":"https://github.com/matrix-asc/dela","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"PCP-MAE","metrics":{"OBJ-BG (OA)":"95.52","OBJ-ONLY (OA)":"94.32","Overall Accuracy":"90.35"},"uses_additional_data":false,"paper_date":"2024-08-16","paper":"/paper/pcp-mae-learning-to-predict-centers-for-point","paper_url":"https://arxiv.org/abs/2408.08753v2","paper_title":"PCP-MAE: Learning to Predict Centers for Point Masked Autoencoders","code":"https://github.com/aHapBean/PCP-MAE","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"PointConT","metrics":{"Mean Accuracy":"88.5","Overall Accuracy":"90.3"},"uses_additional_data":false,"paper_date":"2023-03-08","paper":"/paper/point-cloud-classification-using-content","paper_url":"https://arxiv.org/abs/2303.04599v1","paper_title":"Point Cloud Classification Using Content-based Transformer via Clustering in Feature Space","code":"https://github.com/yahuiliu99/pointcont","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"Point-RAE (no voting)","metrics":{"OBJ-BG (OA)":"95.53","OBJ-ONLY (OA)":"93.63","Overall Accuracy":"90.28"},"uses_additional_data":true,"paper_date":"2023-09-25","paper":"/paper/regress-before-construct-regress-autoencoder","paper_url":"https://arxiv.org/abs/2310.03670v1","paper_title":"Regress Before Construct: Regress Autoencoder for Point Cloud Self-supervised Learning","code":"https://github.com/liuyyy111/point-rae","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"Point-FEMAE","metrics":{"OBJ-BG (OA)":"95.18","OBJ-ONLY (OA)":"93.29","Overall Accuracy":"90.22"},"uses_additional_data":true,"paper_date":"2023-12-17","paper":"/paper/towards-compact-3d-representations-via-point","paper_url":"https://arxiv.org/abs/2312.10726v1","paper_title":"Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders","code":"https://github.com/zyh16143998882/aaai24-pointfemae","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"I2P-MAE (no voting)","metrics":{"OBJ-BG (OA)":"94.15","OBJ-ONLY (OA)":"91.57","Overall Accuracy":"90.11"},"uses_additional_data":true,"paper_date":"2022-12-13","paper":"/paper/learning-3d-representations-from-2d-pre","paper_url":"https://arxiv.org/abs/2212.06785v1","paper_title":"Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders","code":"https://github.com/zrrskywalker/i2p-mae","n_code_links":2,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"ULIP + PointNeXt","metrics":{"Mean Accuracy":"88.6","Number of params":"1.4M","Overall Accuracy":"89.7"},"uses_additional_data":true,"paper_date":"2022-12-10","paper":"/paper/ulip-learning-unified-representation-of","paper_url":"https://arxiv.org/abs/2212.05171v4","paper_title":"ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding","code":"https://github.com/salesforce/ulip","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"ReCon+PPT","metrics":{"OBJ-BG (OA)":"95.01","OBJ-ONLY (OA)":"93.28","Overall Accuracy":"89.52"},"uses_additional_data":false,"paper_date":"2024-08-21","paper":"/paper/positional-prompt-tuning-for-efficient-3d","paper_url":"https://arxiv.org/abs/2408.11567v1","paper_title":"Positional Prompt Tuning for Efficient 3D Representation Learning","code":"https://github.com/zsc000722/ppt","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"3D-JEPA","metrics":{"OBJ-BG (OA)":"93.63","OBJ-ONLY (OA)":"94.49","Overall Accuracy":"89.52"},"uses_additional_data":true,"paper_date":"2024-09-24","paper":"/paper/3d-jepa-a-joint-embedding-predictive","paper_url":"https://arxiv.org/abs/2409.15803v1","paper_title":"3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":25,"model":"PointMLP∗ + JM3D","metrics":{"Mean Accuracy":"88.7","Overall Accuracy":"89.5"},"uses_additional_data":true,"paper_date":"2023-08-06","paper":"/paper/beyond-first-impressions-integrating-joint","paper_url":"https://arxiv.org/abs/2308.02982v2","paper_title":"Beyond First Impressions: Integrating Joint Multi-modal Cues for Comprehensive 3D Representation","code":"https://github.com/mr-neko/jm3d","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"ULIP + PointMLP","metrics":{"Mean Accuracy":"88.5","Overall Accuracy":"89.4"},"uses_additional_data":true,"paper_date":"2022-12-10","paper":"/paper/ulip-learning-unified-representation-of","paper_url":"https://arxiv.org/abs/2212.05171v4","paper_title":"ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding","code":"https://github.com/salesforce/ulip","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"KPConvX-L","metrics":{"Mean Accuracy":"88.1","Overall Accuracy":"89.3"},"uses_additional_data":false,"paper_date":"2024-05-21","paper":"/paper/kpconvx-modernizing-kernel-point-convolution","paper_url":"https://arxiv.org/abs/2405.13194v1","paper_title":"KPConvX: Modernizing Kernel Point Convolution with Kernel Attention","code":"https://github.com/apple/ml-kpconvx","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"P2P","metrics":{"Number of params":"195.8M","Overall Accuracy":"89.3"},"uses_additional_data":true,"paper_date":"2022-08-04","paper":"/paper/p2p-tuning-pre-trained-image-models-for-point","paper_url":"https://arxiv.org/abs/2208.02812v2","paper_title":"P2P: Tuning Pre-trained Image Models for Point Cloud Analysis with Point-to-Pixel Prompting","code":"https://github.com/wangzy22/P2P","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":9,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"ACT","metrics":{"Overall Accuracy":"89.17"},"uses_additional_data":true,"paper_date":"2022-12-16","paper":"/paper/autoencoders-as-cross-modal-teachers-can","paper_url":"https://arxiv.org/abs/2212.08320v2","paper_title":"Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?","code":"https://github.com/qizekun/ReCon","n_code_links":4,"syntology":null},{"rank_in_archive_order":30,"model":"Ours","metrics":{"Mean Accuracy":"87.2","Overall Accuracy":"89.0"},"uses_additional_data":false,"paper_date":"2024-11-01","paper":"/paper/beyond-local-patches-preserving-global-local","paper_url":"https://www.sciencedirect.com/science/article/pii/S0031320324004631","paper_title":"Beyond local patches: Preserving global–local interactions by enhancing self-attention via 3D point cloud tokenization","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":31,"model":"OTMae3D","metrics":{"FLOPs":"6.29","OBJ-BG (OA)":"92.9","OBJ-ONLY (OA)":"92.3","Overall Accuracy":"89.0"},"uses_additional_data":true,"paper_date":"2024-12-26","paper":"/paper/rethinking-masked-representation-learning-for","paper_url":"https://ieeexplore.ieee.org/document/10815033","paper_title":"Rethinking Masked Representation Learning for 3D Point Cloud Understanding","code":"https://github.com/OpenSpaceAI/OTMae3D","n_code_links":1,"syntology":null},{"rank_in_archive_order":32,"model":"ULIP-2 + Point-BERT","metrics":{"Overall Accuracy":"89.0"},"uses_additional_data":true,"paper_date":"2023-05-14","paper":"/paper/ulip-2-towards-scalable-multimodal-pre","paper_url":"https://arxiv.org/abs/2305.08275v4","paper_title":"ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding","code":"https://github.com/salesforce/ulip","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":33,"model":"PointNeXt+Local","metrics":{"Mean Accuracy":"87.4","Overall Accuracy":"88.6"},"uses_additional_data":false,"paper_date":"2022-12-09","paper":"/paper/local-neighborhood-features-for-3d-1","paper_url":"https://arxiv.org/abs/2212.05140v2","paper_title":"Local Neighborhood Features for 3D Classification","code":"https://github.com/VimsLab/Local3DFeatures","n_code_links":1,"syntology":null},{"rank_in_archive_order":34,"model":"SPoTr","metrics":{"Mean Accuracy":"86.8","Overall Accuracy":"88.6"},"uses_additional_data":false,"paper_date":"2023-03-29","paper":"/paper/self-positioning-point-based-transformer-for","paper_url":"https://arxiv.org/abs/2303.16450v1","paper_title":"Self-positioning Point-based Transformer for Point Cloud Understanding","code":"https://github.com/mlvlab/spotr","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":11,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":35,"model":"IDPT","metrics":{"OBJ-BG (OA)":"93.12","Overall Accuracy":"88.51"},"uses_additional_data":false,"paper_date":"2023-04-14","paper":"/paper/instance-aware-dynamic-prompt-tuning-for-pre","paper_url":"https://arxiv.org/abs/2304.07221v2","paper_title":"Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Models","code":"https://github.com/zyh16143998882/iccv23-idpt","n_code_links":3,"syntology":{"n_ran":9,"n_unverified":5,"n_samples":14,"n_pointer_only_licence":14}},{"rank_in_archive_order":36,"model":"PointMLP+TAP","metrics":{"Overall Accuracy":"88.5"},"uses_additional_data":false,"paper_date":"2023-07-27","paper":"/paper/take-a-photo-3d-to-2d-generative-pre-training","paper_url":"https://arxiv.org/abs/2307.14971v2","paper_title":"Take-A-Photo: 3D-to-2D Generative Pre-training of Point Cloud Models","code":"https://github.com/wangzy22/tap","n_code_links":1,"syntology":null},{"rank_in_archive_order":37,"model":"PointNeXt+GAM","metrics":{"FLOPs":"1.64G","Number of params":"1.4M","Overall Accuracy":"88.4"},"uses_additional_data":false,"paper_date":"2015-12-14","paper":null,"paper_url":"http://arxiv.org/abs/1505.07110v2","paper_title":"$(0, 4)$ dualities","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":38,"model":"PointNeXt+HyCoRe","metrics":{"Mean Accuracy":"87.0","Overall Accuracy":"88.3"},"uses_additional_data":false,"paper_date":"2022-09-21","paper":"/paper/rethinking-the-compositionality-of-point","paper_url":"https://arxiv.org/abs/2209.10318v1","paper_title":"Rethinking the compositionality of point clouds through regularization in the hyperbolic space","code":"https://github.com/diegovalsesia/hycore","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":39,"model":"ACT (no voting)","metrics":{"OBJ-BG (OA)":"93.29","OBJ-ONLY (OA)":"91.91","Overall Accuracy":"88.21"},"uses_additional_data":true,"paper_date":"2022-12-16","paper":"/paper/autoencoders-as-cross-modal-teachers-can","paper_url":"https://arxiv.org/abs/2212.08320v2","paper_title":"Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?","code":"https://github.com/qizekun/ReCon","n_code_links":4,"syntology":null},{"rank_in_archive_order":40,"model":"PointNeXt","metrics":{"FLOPs":"1.64G","Mean Accuracy":"86.8","Number of params":"1.4M","Overall Accuracy":"88.2"},"uses_additional_data":false,"paper_date":"2022-06-09","paper":"/paper/pointnext-revisiting-pointnet-with-improved","paper_url":"https://arxiv.org/abs/2206.04670v2","paper_title":"PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies","code":"https://github.com/guochengqian/pointnext","n_code_links":3,"syntology":{"n_ran":7,"n_unverified":10,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":41,"model":"PointConT (no voting)","metrics":{"Mean Accuracy":"86.0","Overall Accuracy":"88.0"},"uses_additional_data":false,"paper_date":"2023-03-08","paper":"/paper/point-cloud-classification-using-content","paper_url":"https://arxiv.org/abs/2303.04599v1","paper_title":"Point Cloud Classification Using Content-based Transformer via Clustering in Feature Space","code":"https://github.com/yahuiliu99/pointcont","n_code_links":1,"syntology":null},{"rank_in_archive_order":42,"model":"PointVector-S","metrics":{"Mean Accuracy":"86.2","Overall Accuracy":"87.8"},"uses_additional_data":false,"paper_date":"2022-05-21","paper":"/paper/point-is-a-vector-a-feature-representation-in","paper_url":"https://arxiv.org/abs/2205.10528v3","paper_title":"PointVector: A Vector Representation In Point Cloud Analysis","code":"https://github.com/guochengqian/openpoints/blob/master/models/backbone/pointvector.py","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":43,"model":"point2vec","metrics":{"Mean Accuracy":"86.0","OBJ-BG (OA)":"91.2","OBJ-ONLY (OA)":"90.4","Overall Accuracy":"87.5"},"uses_additional_data":true,"paper_date":"2023-03-29","paper":"/paper/point2vec-for-self-supervised-representation","paper_url":"https://arxiv.org/abs/2303.16570v2","paper_title":"Point2Vec for Self-Supervised Representation Learning on Point Clouds","code":"https://github.com/kabouzeid/point2vec","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":44,"model":"PointStack","metrics":{"Mean Accuracy":"86.2","Overall Accuracy":"87.2"},"uses_additional_data":false,"paper_date":"2022-05-20","paper":"/paper/advanced-feature-learning-on-point-clouds","paper_url":"https://arxiv.org/abs/2205.09962v1","paper_title":"Advanced Feature Learning on Point Clouds using Multi-resolution Features and Learnable Pooling","code":"https://github.com/LongerVision/PointStack","n_code_links":2,"syntology":null},{"rank_in_archive_order":45,"model":"Point-PN","metrics":{"Number of params":"0.8M","Overall Accuracy":"87.1"},"uses_additional_data":false,"paper_date":"2023-03-14","paper":"/paper/parameter-is-not-all-you-need-starting-from","paper_url":"https://arxiv.org/abs/2303.08134v2","paper_title":"Parameter is Not All You Need: Starting from Non-Parametric Networks for 3D Point Cloud Analysis","code":"https://github.com/zrrskywalker/point-nn","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":46,"model":"PointCMT","metrics":{"Mean Accuracy":"84.8","Number of params":"12.6M","Overall Accuracy":"86.7"},"uses_additional_data":false,"paper_date":"2022-10-09","paper":"/paper/let-images-give-you-more-point-cloud-cross","paper_url":"https://arxiv.org/abs/2210.04208v1","paper_title":"Let Images Give You More:Point Cloud Cross-Modal Training for Shape Analysis","code":"https://github.com/yanx27/2dpass","n_code_links":2,"syntology":null},{"rank_in_archive_order":47,"model":"Point-JEPA","metrics":{"OBJ-BG (OA)":"92.9±0.4","Overall Accuracy":"86.6"},"uses_additional_data":true,"paper_date":"2024-04-25","paper":"/paper/point-jepa-a-joint-embedding-predictive","paper_url":"https://arxiv.org/abs/2404.16432v5","paper_title":"Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud","code":"https://github.com/Ayumu-J-S/Point-JEPA","n_code_links":1,"syntology":null},{"rank_in_archive_order":48,"model":"PointMLS","metrics":{"Overall Accuracy":"86.6"},"uses_additional_data":false,"paper_date":"2024-01-16","paper":"/paper/modelnet-o-a-large-scale-synthetic-dataset","paper_url":"https://arxiv.org/abs/2401.08210v1","paper_title":"ModelNet-O: A Large-Scale Synthetic Dataset for Occlusion-Aware Point Cloud Classification","code":"https://github.com/fanglaosi/pointmls","n_code_links":1,"syntology":null},{"rank_in_archive_order":49,"model":"Point-M2AE","metrics":{"OBJ-BG (OA)":"91.22","OBJ-ONLY (OA)":"88.81","Overall Accuracy":"86.43"},"uses_additional_data":true,"paper_date":"2022-05-28","paper":"/paper/point-m2ae-multi-scale-masked-autoencoders","paper_url":"https://arxiv.org/abs/2205.14401v2","paper_title":"Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training","code":"https://github.com/zrrskywalker/i2p-mae","n_code_links":3,"syntology":null},{"rank_in_archive_order":50,"model":"ULIP + PointBERT","metrics":{"Overall Accuracy":"86.4"},"uses_additional_data":true,"paper_date":"2022-12-10","paper":"/paper/ulip-learning-unified-representation-of","paper_url":"https://arxiv.org/abs/2212.05171v4","paper_title":"ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding","code":"https://github.com/salesforce/ulip","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":51,"model":"DualMLP","metrics":{"Overall Accuracy":"86.4"},"uses_additional_data":false,"paper_date":"2023-10-10","paper":"/paper/dualmlp-a-two-stream-fusion-model-for-3d","paper_url":"https://link.springer.com/article/10.1007/s00371-023-03114-3","paper_title":"DualMLP: a two-stream fusion model for 3D point cloud classification","code":"https://github.com/snehaputul/DualMLP","n_code_links":1,"syntology":null},{"rank_in_archive_order":52,"model":"RepSurf-U (2x)","metrics":{"FLOPs":"2.43G","Number of params":"6.80M","Overall Accuracy":"86.0"},"uses_additional_data":false,"paper_date":"2022-05-11","paper":"/paper/surface-representation-for-point-clouds","paper_url":"https://arxiv.org/abs/2205.05740v2","paper_title":"Surface Representation for Point Clouds","code":"https://github.com/hancyran/RepSurf","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":13,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":53,"model":"PointMLP","metrics":{"Mean Accuracy":"84.4","Overall Accuracy":"85.7"},"uses_additional_data":false,"paper_date":"2022-02-15","paper":"/paper/rethinking-network-design-and-local-geometry-1","paper_url":"https://arxiv.org/abs/2202.07123v2","paper_title":"Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework","code":"https://github.com/ma-xu/pointmlp-pytorch","n_code_links":1,"syntology":{"n_ran":10,"n_unverified":0,"n_samples":10,"n_pointer_only_licence":2}},{"rank_in_archive_order":54,"model":"Point-LGMask","metrics":{"OBJ-BG (OA)":"89.8","OBJ-ONLY (OA)":"89.3","Overall Accuracy":"85.3"},"uses_additional_data":true,"paper_date":"2023-06-08","paper":"/paper/point-lgmask-local-and-global-contexts","paper_url":"https://ieeexplore.ieee.org/abstract/document/10146493","paper_title":"Point-LGMask: Local and Global Contexts Embedding for Point Cloud Pre-training with Multi-Ratio Masking","code":"https://github.com/TangYuan96/Point-LGMask","n_code_links":1,"syntology":null},{"rank_in_archive_order":55,"model":"Point-MAE","metrics":{"OBJ-BG (OA)":"90.02","OBJ-ONLY (OA)":"88.29","Overall Accuracy":"85.2"},"uses_additional_data":true,"paper_date":"2022-03-13","paper":"/paper/masked-autoencoders-for-point-cloud-self","paper_url":"https://arxiv.org/abs/2203.06604v2","paper_title":"Masked Autoencoders for Point Cloud Self-supervised Learning","code":"https://github.com/Pang-Yatian/Point-MAE","n_code_links":4,"syntology":{"n_ran":7,"n_unverified":2,"n_samples":9,"n_pointer_only_licence":1}},{"rank_in_archive_order":56,"model":"MVTN+SimpleView++","metrics":{"Overall Accuracy":"84.8"},"uses_additional_data":false,"paper_date":"2022-09-08","paper":"/paper/simpleview-neighborhood-views-for-point-cloud","paper_url":"https://ieeexplore.ieee.org/document/9874679","paper_title":"SimpleView++: Neighborhood Views for Point Cloud Classification","code":"https://github.com/VimsLab/SimpleViewPlusPlus","n_code_links":1,"syntology":null},{"rank_in_archive_order":57,"model":"DeltaConv","metrics":{"Overall Accuracy":"84.7"},"uses_additional_data":false,"paper_date":"2021-11-16","paper":"/paper/deltaconv-anisotropic-point-cloud-learning","paper_url":"https://arxiv.org/abs/2111.08799v5","paper_title":"DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds","code":"https://github.com/rubenwiersma/deltaconv","n_code_links":1,"syntology":null},{"rank_in_archive_order":58,"model":"RepSurf-U","metrics":{"FLOPs":"0.81G","Number of params":"1.48M","Overall Accuracy":"84.6"},"uses_additional_data":false,"paper_date":"2022-05-11","paper":"/paper/surface-representation-for-point-clouds","paper_url":"https://arxiv.org/abs/2205.05740v2","paper_title":"Surface Representation for Point Clouds","code":"https://github.com/hancyran/RepSurf","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":13,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":59,"model":"APP-Net","metrics":{"Overall Accuracy":"84.1"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/app-net-auxiliary-point-based-push-and-pull","paper_url":"https://arxiv.org/abs/2205.00847v2","paper_title":"APP-Net: Auxiliary-point-based Push and Pull Operations for Efficient Point Cloud Classification","code":"https://github.com/mcg-nju/app-net","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":60,"model":"PointMLP-elite","metrics":{"Mean Accuracy":"81.8","Overall Accuracy":" 83.8"},"uses_additional_data":false,"paper_date":"2022-02-15","paper":"/paper/rethinking-network-design-and-local-geometry-1","paper_url":"https://arxiv.org/abs/2202.07123v2","paper_title":"Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework","code":"https://github.com/ma-xu/pointmlp-pytorch","n_code_links":1,"syntology":{"n_ran":10,"n_unverified":0,"n_samples":10,"n_pointer_only_licence":2}},{"rank_in_archive_order":61,"model":"PointNet++ + SageMix","metrics":{"Overall Accuracy":"83.7"},"uses_additional_data":false,"paper_date":"2022-10-13","paper":"/paper/sagemix-saliency-guided-mixup-for-point","paper_url":"https://arxiv.org/abs/2210.06944v1","paper_title":"SageMix: Saliency-Guided Mixup for Point Clouds","code":"https://github.com/mlvlab/SageMix","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":62,"model":"DGCNN + SageMix","metrics":{"Overall Accuracy":"83.6"},"uses_additional_data":false,"paper_date":"2022-10-13","paper":"/paper/sagemix-saliency-guided-mixup-for-point","paper_url":"https://arxiv.org/abs/2210.06944v1","paper_title":"SageMix: Saliency-Guided Mixup for Point Clouds","code":"https://github.com/mlvlab/SageMix","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":63,"model":"Point-TnT","metrics":{"FLOPs":"1.19G","Mean Accuracy":"81.0","Number of params":"3.9M","Overall Accuracy":"83.5"},"uses_additional_data":false,"paper_date":"2022-04-08","paper":"/paper/points-to-patches-enabling-the-use-of-self","paper_url":"https://arxiv.org/abs/2204.03957v1","paper_title":"Points to Patches: Enabling the Use of Self-Attention for 3D Shape Recognition","code":"https://github.com/axeber01/point-tnt","n_code_links":1,"syntology":null},{"rank_in_archive_order":64,"model":"Point-BERT","metrics":{"OBJ-BG (OA)":"87.43","OBJ-ONLY (OA)":"88.12","Overall Accuracy":"83.1"},"uses_additional_data":true,"paper_date":"2021-11-29","paper":"/paper/point-bert-pre-training-3d-point-cloud","paper_url":"https://arxiv.org/abs/2111.14819v2","paper_title":"Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modeling","code":"https://github.com/julie-tang00/point-bert","n_code_links":3,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":65,"model":"MVTN","metrics":{"OBJ-BG (OA)":"92.6","OBJ-ONLY (OA)":"92.3","Overall Accuracy":"82.8"},"uses_additional_data":false,"paper_date":"2020-11-26","paper":"/paper/mvtn-multi-view-transformation-network-for-3d","paper_url":"https://arxiv.org/abs/2011.13244v3","paper_title":"MVTN: Multi-View Transformation Network for 3D Shape Recognition","code":"https://github.com/ajhamdi/MVTN","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":66,"model":"PRA-Net","metrics":{"Mean Accuracy":"79.1","Overall Accuracy":"82.1"},"uses_additional_data":false,"paper_date":"2021-12-09","paper":"/paper/pra-net-point-relation-aware-network-for-3d","paper_url":"https://arxiv.org/abs/2112.04903v1","paper_title":"PRA-Net: Point Relation-Aware Network for 3D Point Cloud Analysis","code":"https://github.com/xiwuchen/pra-net","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":67,"model":"DynamicScale","metrics":{"Overall Accuracy":"82.0"},"uses_additional_data":false,"paper_date":"2021-10-19","paper":"/paper/dynamic-local-geometry-capture-in-3d","paper_url":"https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9565556","paper_title":"Dynamic Local Geometry Capture in 3D PointCloud Classification","code":"https://github.com/VimsLab/DynamicScale","n_code_links":1,"syntology":null},{"rank_in_archive_order":68,"model":"PatchAugment","metrics":{"Mean Accuracy":"79.7","Overall Accuracy":"81.0"},"uses_additional_data":false,"paper_date":"2021-10-19","paper":"/paper/patchaugment-local-neighborhood-augmentation","paper_url":"https://openaccess.thecvf.com/content/ICCV2021W/DLGC/html/Sheshappanavar_PatchAugment_Local_Neighborhood_Augmentation_in_Point_Cloud_Classification_ICCVW_2021_paper.html","paper_title":"PatchAugment: Local Neighborhood Augmentation in Point Cloud Classification","code":"https://github.com/VimsLab/PatchAugment","n_code_links":1,"syntology":null},{"rank_in_archive_order":69,"model":"GBNet","metrics":{"Mean Accuracy":"77.8","Overall Accuracy":"80.5"},"uses_additional_data":false,"paper_date":"2019-11-28","paper":"/paper/geometric-feedback-network-for-point-cloud","paper_url":"https://arxiv.org/abs/1911.12885v5","paper_title":"Geometric Back-projection Network for Point Cloud Classification","code":"https://github.com/ShiQiu0419/GFNet","n_code_links":2,"syntology":null},{"rank_in_archive_order":70,"model":"SimpleView","metrics":{"Overall Accuracy":"80.5"},"uses_additional_data":false,"paper_date":"2021-01-01","paper":"/paper/revisiting-point-cloud-classification-with-a","paper_url":"https://openreview.net/forum?id=XwATtbX3oCz","paper_title":"Revisiting Point Cloud Classification with a Simple and Effective Baseline","code":"https://github.com/princeton-vl/SimpleView","n_code_links":2,"syntology":null},{"rank_in_archive_order":71,"model":"DRNet","metrics":{"Mean Accuracy":"78.0","Overall Accuracy":"80.3"},"uses_additional_data":false,"paper_date":"2020-05-14","paper":"/paper/dense-resolution-network-for-point-cloud","paper_url":"https://arxiv.org/abs/2005.06734v2","paper_title":"Dense-Resolution Network for Point Cloud Classification and Segmentation","code":"https://github.com/ShiQiu0419/DRNet","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":5,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":72,"model":"PointCNN","metrics":{"Mean Accuracy":"75.1","OBJ-BG (OA)":"86.1","OBJ-ONLY (OA)":"85.5","Overall Accuracy":"78.5"},"uses_additional_data":false,"paper_date":"2018-01-23","paper":"/paper/pointcnn-convolution-on-mathcalx-transformed","paper_url":"http://arxiv.org/abs/1801.07791v5","paper_title":"PointCNN: Convolution On $\\mathcal{X}$-Transformed Points","code":"https://github.com/pyg-team/pytorch_geometric","n_code_links":16,"syntology":null},{"rank_in_archive_order":73,"model":"DGCNN","metrics":{"Mean Accuracy":"73.6","OBJ-BG (OA)":"82.8","OBJ-ONLY (OA)":"86.2","Overall Accuracy":"78.1"},"uses_additional_data":false,"paper_date":"2018-01-24","paper":"/paper/dynamic-graph-cnn-for-learning-on-point","paper_url":"https://arxiv.org/abs/1801.07829v2","paper_title":"Dynamic Graph CNN for Learning on Point Clouds","code":"https://github.com/WangYueFt/dgcnn","n_code_links":21,"syntology":{"n_ran":16,"n_unverified":28,"n_samples":44,"n_pointer_only_licence":31}},{"rank_in_archive_order":74,"model":"PointNet++","metrics":{"Mean Accuracy":"75.4","OBJ-BG (OA)":"82.3","OBJ-ONLY (OA)":"84.3","Overall Accuracy":"77.9"},"uses_additional_data":false,"paper_date":"2017-06-07","paper":"/paper/pointnet-deep-hierarchical-feature-learning","paper_url":"http://arxiv.org/abs/1706.02413v1","paper_title":"PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space","code":"https://github.com/yanx27/Pointnet_Pointnet2_pytorch","n_code_links":68,"syntology":{"n_ran":36,"n_unverified":31,"n_samples":67,"n_pointer_only_licence":26}},{"rank_in_archive_order":75,"model":"SpiderCNN","metrics":{"Mean Accuracy":"69.8","Overall Accuracy":"73.7"},"uses_additional_data":false,"paper_date":"2018-03-30","paper":"/paper/spidercnn-deep-learning-on-point-sets-with","paper_url":"http://arxiv.org/abs/1803.11527v3","paper_title":"SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filters","code":"https://github.com/xyf513/SpiderCNN","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":6,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":76,"model":"PointNet","metrics":{"Mean Accuracy":"63.4","Overall Accuracy":"68.2"},"uses_additional_data":false,"paper_date":"2016-12-02","paper":"/paper/pointnet-deep-learning-on-point-sets-for-3d","paper_url":"http://arxiv.org/abs/1612.00593v2","paper_title":"PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation","code":"https://github.com/charlesq34/pointnet","n_code_links":110,"syntology":{"n_ran":89,"n_unverified":75,"n_samples":164,"n_pointer_only_licence":90}},{"rank_in_archive_order":77,"model":"ExpPoint-MAE","metrics":{"OBJ-BG (OA)":"90.88","OBJ-ONLY (OA)":"90.02"},"uses_additional_data":true,"paper_date":"2023-06-19","paper":"/paper/a-deep-dive-into-explainable-self-supervised","paper_url":"https://arxiv.org/abs/2306.10798v3","paper_title":"ExpPoint-MAE: Better interpretability and performance for self-supervised point cloud transformers","code":"https://github.com/vvrpanda/exppoint-mae","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":38,"rows_with_any_sample_ran":36,"distinct_papers_with_graph_line":29,"distinct_papers_with_any_sample_ran":28,"samples_over_distinct_papers":{"n_ran":237,"n_unverified":229,"n_samples":466,"n_pointer_only_licence":178,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":261,"n_unverified":247,"n_samples":508,"n_pointer_only_licence":189,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}