Escaping Plato's Cave: Towards the Alignment of 3D and Text Latent Spaces

• URL: http://arxiv.org/abs/2503.05283v1
• Date: Fri, 07 Mar 2025 09:51:56 GMT
• Title: Escaping Plato's Cave: Towards the Alignment of 3D and Text Latent Spaces
• Authors: Souhail Hadgi, Luca Moschella, Andrea Santilli, Diego Gomez, Qixing Huang, Emanuele Rodolà, Simone Melzi, Maks Ovsjanikov,
• Abstract summary: We show that naive post-training feature alignment of uni-modal text and 3D encoders results in limited performance.<n>We then focus on extracting subspaces of the corresponding feature spaces and discover that by projecting learned representations onto well-chosen lower-dimensional subspaces the quality of alignment becomes significantly higher.
• Score: 52.237827968294766
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent works have shown that, when trained at scale, uni-modal 2D vision and text encoders converge to learned features that share remarkable structural properties, despite arising from different representations. However, the role of 3D encoders with respect to other modalities remains unexplored. Furthermore, existing 3D foundation models that leverage large datasets are typically trained with explicit alignment objectives with respect to frozen encoders from other representations. In this work, we investigate the possibility of a posteriori alignment of representations obtained from uni-modal 3D encoders compared to text-based feature spaces. We show that naive post-training feature alignment of uni-modal text and 3D encoders results in limited performance. We then focus on extracting subspaces of the corresponding feature spaces and discover that by projecting learned representations onto well-chosen lower-dimensional subspaces the quality of alignment becomes significantly higher, leading to improved accuracy on matching and retrieval tasks. Our analysis further sheds light on the nature of these shared subspaces, which roughly separate between semantic and geometric data representations. Overall, ours is the first work that helps to establish a baseline for post-training alignment of 3D uni-modal and text feature spaces, and helps to highlight both the shared and unique properties of 3D data compared to other representations.

Related papers

• Enhanced Cross-modal 3D Retrieval via Tri-modal Reconstruction [4.820576346277399]
  Cross-modal 3D retrieval is a critical yet challenging task, aiming to achieve bi-directional retrieval between 3D and text modalities. We propose to adopt multi-view images and point clouds to jointly represent 3D shapes, facilitating tri-modal alignment. Our method significantly outperforms previous state-of-the-art methods in both shape-to-text and text-to-shape retrieval tasks.
  arXiv  Detail & Related papers  (2025-04-02T08:29:42Z)
• From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3D [32.547597353581594]
  We introduce a novel 2D spatial data generation and annotation pipeline built upon scene data with 3D ground-truth. We construct SPAR-7M, a large-scale dataset generated from thousands of scenes across multiple public datasets. In addition, we introduce SPAR-Bench, a benchmark designed to offer a more comprehensive evaluation of spatial capabilities.
  arXiv  Detail & Related papers  (2025-03-29T04:51:50Z)
• 3D Weakly Supervised Semantic Segmentation with 2D Vision-Language Guidance [68.8825501902835]
  3DSS-VLG is a weakly supervised approach for 3D Semantic with 2D Vision-Language Guidance. To the best of our knowledge, this is the first work to investigate 3D weakly supervised semantic segmentation by using the textual semantic information of text category labels.
  arXiv  Detail & Related papers  (2024-07-13T09:39:11Z)
• Enhancing Generalizability of Representation Learning for Data-Efficient 3D Scene Understanding [50.448520056844885]
  We propose a generative Bayesian network to produce diverse synthetic scenes with real-world patterns. A series of experiments robustly display our method's consistent superiority over existing state-of-the-art pre-training approaches.
  arXiv  Detail & Related papers  (2024-06-17T07:43:53Z)
• MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations [55.022519020409405]
  This paper builds the first largest ever multi-modal 3D scene dataset and benchmark with hierarchical grounded language annotations, MMScan. The resulting multi-modal 3D dataset encompasses 1.4M meta-annotated captions on 109k objects and 7.7k regions as well as over 3.04M diverse samples for 3D visual grounding and question-answering benchmarks.
  arXiv  Detail & Related papers  (2024-06-13T17:59:30Z)
• DatasetNeRF: Efficient 3D-aware Data Factory with Generative Radiance Fields [68.94868475824575]
  This paper introduces a novel approach capable of generating infinite, high-quality 3D-consistent 2D annotations alongside 3D point cloud segmentations. We leverage the strong semantic prior within a 3D generative model to train a semantic decoder. Once trained, the decoder efficiently generalizes across the latent space, enabling the generation of infinite data.
  arXiv  Detail & Related papers  (2023-11-18T21:58:28Z)
• CLIP$^2$: Contrastive Language-Image-Point Pretraining from Real-World Point Cloud Data [80.42480679542697]
  We propose Contrastive Language-Image-Point Cloud Pretraining (CLIP$2$) to learn the transferable 3D point cloud representation in realistic scenarios. Specifically, we exploit naturally-existed correspondences in 2D and 3D scenarios, and build well-aligned and instance-based text-image-point proxies from those complex scenarios.
  arXiv  Detail & Related papers  (2023-03-22T09:32:45Z)
• Pri3D: Can 3D Priors Help 2D Representation Learning? [37.35721274841419]
  We introduce an approach to learn view-invariant,geometry-aware representations for network pre-training. We employ contrastive learning under both multi-view im-age constraints and image-geometry constraints to encode3D priors into learned 2D representations.
  arXiv  Detail & Related papers  (2021-04-22T17:59:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.