NeRF-UAVeL: Unified Attention-driven Volumetric Learning for Enhanced NeRF-based 3D Object Detection

Three-dimensional (3D) object detection based on Neural Radiance Fields (NeRF) has emerged as a promising direction for reconstructing complex environments from posed RGB images. However, existing NeRF-based detectors often suffer from coarse feature encoding and limited attention to multi-scale volumetric structure, leading to inaccurate localization and poor generalization in real-world scenarios. NeRF-UAVeL addresses these challenges with a unified attention-driven volumetric learning detection framework that integrates four novel modules into a NeRF-derived 3D volumetric backbone: Multi-dimensional Volumetric Attention Pooling (MVAP), Tri-Scale Asymmetric Convolutional Aggregation (TACA), Dual-Domain Attention Fusion (DDAF), and Volumetric Cross-Window Attention Fusion (V-CWAF). Extensive experiments on the 3D-FRONT and ScanNet datasets demonstrate that NeRF-UAVeL outperforms both point cloud-based and multi-view-based methods, improving AP50 by +6.7% and R50 by +7.3% over the baseline on 3D-FRONT, and achieving +6.9% in AP50 and +2.9% in R50 on ScanNet.