Gen-LangSplat：基于预训练特征压缩的广义语言高斯泼溅
Gen-LangSplat: Generalized Language Gaussian Splatting with Pre-Trained Feature Compression
AI总结：
Gen-LangSplat提出用预训练于ScanNet的广义自编码器替代逐场景自编码器，实现无需场景特定训练的高效3D开放词汇语言场构建，性能媲美或超越LangSplat。
摘要：
在3D空间中建模开放词汇语言场对于实现直观的人机交互以及在物理环境中的查询至关重要。最先进的方法（如LangSplat）利用3D高斯泼溅高效地构建这些语言场，编码从高维模型（如CLIP）中蒸馏出的特征。然而，这种效率目前被需要训练场景特定的语言自编码器进行特征压缩所抵消，这引入了昂贵的逐场景优化瓶颈，阻碍了部署的可扩展性。在本工作中，我们提出了Gen-LangSplat，通过用在大规模ScanNet数据集上广泛预训练的广义自编码器替换逐场景自编码器，消除了这一需求。这种架构转变使得任何新场景都能使用固定的、紧凑的语言特征潜空间，而无需任何场景特定的训练。通过消除这一依赖，我们的整个语言场构建过程实现了效率提升，同时提供了与原始LangSplat方法相当或更优的查询性能。为了验证我们的设计选择，我们进行了彻底的消融研究，经验性地确定了最优潜嵌入维度，并使用均方误差和余弦相似度量化了原始与重投影的512维CLIP嵌入之间的表示保真度。我们的结果表明，广义嵌入能够高效且准确地支持新3D场景中的开放词汇查询，为可扩展的实时交互式3D AI应用铺平了道路。
英文摘要：
Modeling open-vocabulary language fields in 3D is essential for intuitive human-AI interaction and querying within physical environments. State-of-the-art approaches, such as LangSplat, leverage 3D Gaussian Splatting to efficiently construct these language fields, encoding features distilled from high-dimensional models like CLIP. However, this efficiency is currently offset by the requirement to train a scene-specific language autoencoder for feature compression, introducing a costly, per-scene optimization bottleneck that hinders deployment scalability. In this work, we introduce Gen-LangSplat, that eliminates this requirement by replacing the scene-wise autoencoder with a generalized autoencoder, pre-trained extensively on the large-scale ScanNet dataset. This architectural shift enables the use of a fixed, compact latent space for language features across any new scene without any scene-specific training. By removing this dependency, our entire language field construction process achieves a efficiency boost while delivering querying performance comparable to, or exceeding, the original LangSplat method. To validate our design choice, we perform a thorough ablation study empirically determining the optimal latent embedding dimension and quantifying representational fidelity using Mean Squared Error and cosine similarity between the original and reprojected 512-dimensional CLIP embeddings. Our results demonstrate that generalized embeddings can efficiently and accurately support open-vocabulary querying in novel 3D scenes, paving the way for scalable, real-time interactive 3D AI applications.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
