Pith. sign in

REVIEW 5 cited by

Contrastive Gaussian Clustering: Weakly Supervised 3D Scene Segmentation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2404.12784 v1 pith:OFZY42OX submitted 2024-04-19 cs.CV cs.LG

classification cs.CVcs.LG
keywords segmentationmasksgaussianssceneclusteringcontrastivegaussiangenerate
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

We introduce Contrastive Gaussian Clustering, a novel approach capable of provide segmentation masks from any viewpoint and of enabling 3D segmentation of the scene. Recent works in novel-view synthesis have shown how to model the appearance of a scene via a cloud of 3D Gaussians, and how to generate accurate images from a given viewpoint by projecting on it the Gaussians before $\alpha$ blending their color. Following this example, we train a model to include also a segmentation feature vector for each Gaussian. These can then be used for 3D scene segmentation, by clustering Gaussians according to their feature vectors; and to generate 2D segmentation masks, by projecting the Gaussians on a plane and $\alpha$ blending over their segmentation features. Using a combination of contrastive learning and spatial regularization, our method can be trained on inconsistent 2D segmentation masks, and still learn to generate segmentation masks consistent across all views. Moreover, the resulting model is extremely accurate, improving the IoU accuracy of the predicted masks by $+8\%$ over the state of the art. Code and trained models will be released soon.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. DreamGrasp: Zero-Shot 3D Multi-Object Reconstruction from Partial-View Images for Robotic Manipulation

    cs.RO 2025-07 conditional novelty 6.0 of 10

    A zero-shot pipeline reconstructs per-instance 3D geometry in cluttered scenes from two partial RGB views by combining diffusion-based score distillation with learned instance features and text-guided refinement.

  2. Efficient multi-view training for 3D Gaussian Splatting

    cs.CV 2025-06 conditional novelty 6.0 of 10

    Training 3D Gaussian Splatting with multiple images per iteration, using partial rendering and a 3D-aware SSIM loss, improves novel-view synthesis quality over single-view training.

  3. Enhancing LLM Training via Spectral Clipping

    cs.LG 2026-03 unverdicted novelty 5.0 of 10

    SPECTRA improves LLM pretraining via post-clipping of update spectral norms and optional pre-clipping of gradient spikes, framed as Composite Frank-Wolfe regularization.

  4. Hi-LSplat: Hierarchical 3D Language Gaussian Splatting

    cs.CV 2025-06 conditional novelty 5.0 of 10

    Hi-LSplat trains language-augmented 3D Gaussians with a three-level semantic tree and instance/part contrastive losses, improving open-vocabulary 3D segmentation and localization on eight datasets.

  5. The ALMA-QUARKS Survey: III. Clump-to-core fragmentation and search for high-mass starless cores

    astro-ph.GA 2025-08 unverdicted novelty 4.0 of 10

    In 139 infrared-bright massive protoclusters, ALMA resolves 1562 cores whose separations are much smaller than the Jeans length, and finds only two candidate high-mass starless cores.

Pith tools