Papers › PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm

PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm

12 Oct 2023arXiv:2310.08586archive 2025-07-28

Haoyi Zhu, Honghui Yang, Xiaoyang Wu, Di Huang, Sha Zhang, Xianglong He, Hengshuang Zhao, Chunhua Shen, Yu Qiao, Tong He, Wanli Ouyang

In contrast to numerous NLP and 2D vision foundational models, learning a 3D foundational model poses considerably greater challenges. This is primarily due to the inherent data variability and diversity of downstream tasks. In this paper, we introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation, thereby establishing a pathway to 3D foundational models. Considering that informative 3D features should encode rich geometry and appearance cues that can be utilized to render realistic images, we propose to learn 3D representations by differentiable neural rendering. We train a 3D backbone with a devised volumetric neural renderer by comparing the rendered with the real images. Notably, our approach seamlessly integrates the learned 3D encoder into various downstream tasks. These tasks encompass not only high-level challenges such as 3D detection and segmentation but also low-level objectives like 3D reconstruction and image synthesis, spanning both indoor and outdoor scenarios. Besides, we also illustrate the capability of pre-training a 2D backbone using the proposed methodology, surpassing conventional pre-training methods by a large margin. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness. Code and models are available at https://github.com/OpenGVLab/PonderV2.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

OpenGVLab/PonderV2 officialmentioned in papermentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

3D Object Detection3D Reconstruction3D Semantic SegmentationImage GenerationLIDAR Semantic SegmentationNeural RenderingPoint Cloud Pre-trainingSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Semantic Segmentation ScanNet200 PonderV2 + SparseUNet test mIoU 34.6 #11 of 16 Archive leaderboard report
3D Semantic Segmentation ScanNet200 PonderV2 + SparseUNet val mIoU 32.3 #11 of 16 Archive leaderboard report
Semantic Segmentation S3DIS PonderV2 + SparseUNet Mean IoU 79.9 #3 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PonderV2 + SparseUNet mAcc 86.5 #3 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PonderV2 + SparseUNet oAcc 92.5 #3 of 54 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PonderV2 + SparseUNet mAcc 79.0 #14 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PonderV2 + SparseUNet mIoU 73.2 #14 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PonderV2 + SparseUNet oAcc 92.2 #14 of 61 Archive leaderboard report
Semantic Segmentation ScanNet PonderV2 + SparseUNet test mIoU 78.5 #9 of 45 Archive leaderboard report
Semantic Segmentation ScanNet PonderV2 + SparseUNet val mIoU 77.0 #9 of 45 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections