EfficientPENet: Real-Time Depth Completion from Sparse LiDAR via Lightweight Multi-Modal Fusion

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Lopez, Johny J., Ferdaus, Md Meftahul, Abdelguerfi, Mahdi, Netchaev, Anton, Sloan, Steven, Pathak, Ken, Niles, Kendall N.
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911609061900288
author Lopez, Johny J.
Ferdaus, Md Meftahul
Abdelguerfi, Mahdi
Netchaev, Anton
Sloan, Steven
Pathak, Ken
Niles, Kendall N.
author_facet Lopez, Johny J.
Ferdaus, Md Meftahul
Abdelguerfi, Mahdi
Netchaev, Anton
Sloan, Steven
Pathak, Ken
Niles, Kendall N.
contents Depth completion from sparse LiDAR measurements and corresponding RGB images is a prerequisite for accurate 3D perception in robotic systems. Existing methods achieve high accuracy on standard benchmarks but rely on heavy backbone architectures that preclude real-time deployment on embedded hardware. We present EfficientPENet, a two-branch depth completion network that replaces the conventional ResNet encoder with a modernized ConvNeXt backbone, introduces sparsity-invariant convolutions for the depth stream, and refines predictions through a Convolutional Spatial Propagation Network (CSPN). The RGB branch leverages ImageNet-pretrained ConvNeXt blocks with Layer Normalization, 7x7 depthwise convolutions, and stochastic depth regularization. Features from both branches are merged via late fusion and decoded through a multi-scale deep supervision strategy. We further introduce a position-aware test-time augmentation scheme that corrects coordinate tensors during horizontal flipping, yielding consistent error reduction at inference. On the KITTI depth completion benchmark, EfficientPENet achieves an RMSE of 631.94 mm with 36.24M parameters and a latency of 20.51 ms, operating at 48.76 FPS. This represents a 3.7 times reduction in parameters and a 23 times speedup relative to BP-Net, while maintaining competitive accuracy. These results establish EfficientPENet as a practical solution for real-time depth completion on resource-constrained edge platforms such as the NVIDIA Jetson.
format Preprint
id arxiv_https___arxiv_org_abs_2604_18790
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EfficientPENet: Real-Time Depth Completion from Sparse LiDAR via Lightweight Multi-Modal Fusion
Lopez, Johny J.
Ferdaus, Md Meftahul
Abdelguerfi, Mahdi
Netchaev, Anton
Sloan, Steven
Pathak, Ken
Niles, Kendall N.
Computer Vision and Pattern Recognition
Depth completion from sparse LiDAR measurements and corresponding RGB images is a prerequisite for accurate 3D perception in robotic systems. Existing methods achieve high accuracy on standard benchmarks but rely on heavy backbone architectures that preclude real-time deployment on embedded hardware. We present EfficientPENet, a two-branch depth completion network that replaces the conventional ResNet encoder with a modernized ConvNeXt backbone, introduces sparsity-invariant convolutions for the depth stream, and refines predictions through a Convolutional Spatial Propagation Network (CSPN). The RGB branch leverages ImageNet-pretrained ConvNeXt blocks with Layer Normalization, 7x7 depthwise convolutions, and stochastic depth regularization. Features from both branches are merged via late fusion and decoded through a multi-scale deep supervision strategy. We further introduce a position-aware test-time augmentation scheme that corrects coordinate tensors during horizontal flipping, yielding consistent error reduction at inference. On the KITTI depth completion benchmark, EfficientPENet achieves an RMSE of 631.94 mm with 36.24M parameters and a latency of 20.51 ms, operating at 48.76 FPS. This represents a 3.7 times reduction in parameters and a 23 times speedup relative to BP-Net, while maintaining competitive accuracy. These results establish EfficientPENet as a practical solution for real-time depth completion on resource-constrained edge platforms such as the NVIDIA Jetson.
title EfficientPENet: Real-Time Depth Completion from Sparse LiDAR via Lightweight Multi-Modal Fusion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.18790