CLFTv2: Efficient Camera-LiDAR Fusion for Semantic Segmentation via Hierarchical Feature Pyramids

• URL: http://arxiv.org/abs/2609.09881v1
• Date: Wed, 09 Sep 2026 08:37:06 GMT
• Title: CLFTv2: Efficient Camera-LiDAR Fusion for Semantic Segmentation via Hierarchical Feature Pyramids
• Abstract summary: We introduce CLFTv2, a hierarchical camera-LiDAR fusion framework replacing global ViT attention.<n>On ZOD, CLFTv2-Large achieves 53.5% mIoU, improving pedestrian IoU from 35.5% to 44.9% over the prior model.<n>Compared to a Swin-based Mask2Former adaptation, CLFTv2 requires 1.4$times$ fewer GFLOPs and delivers 2.2$times$ higher throughput.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Semantic segmentation for autonomous driving requires reliable detection of vulnerable road users (VRUs) despite heavy class imbalance. We introduce CLFTv2, a hierarchical camera-LiDAR fusion framework replacing global ViT attention with a Swin-based multi-scale encoder and a lightweight FPN-style residual decoder. Operating in the 2D perspective domain, CLFTv2 integrates multi-scale geometric cues through shifted-window attention and per-scale residual fusion, avoiding the computational overhead of query-matching decoders. Across three driving datasets, CLFTv2 consistently improves VRU recall. On ZOD, CLFTv2-Large achieves 53.5\% mIoU, improving pedestrian IoU from 35.5\% to 44.9\% over the prior CLFT model. On Waymo, CLFTv2 reaches 61.7\% mIoU. Additionally, a modality-isolation study suggests ViT's global receptive field yields stronger fusion gains only under dense LiDAR returns. Compared to a Swin-based Mask2Former adaptation, CLFTv2 requires 1.4$\times$ fewer GFLOPs and delivers 2.2$\times$ higher throughput, while achieving comparable overall accuracy. These results demonstrate that hierarchical local-attention fusion offers an efficient, scalable alternative to global-attention and query-based decoders for real-time on-vehicle perception in intelligent transportation systems. Source code is publicly available.

Related papers

• OPUS-V2: Bridging the Gap between Sparse Points and Dense Voxels [74.61043439622689]
  We propose OPUS-V2, a novel framework built upon the pioneering OPUS (occupancy prediction using a sparse set) point-based approach.<n> OPUS-V2 incorporates a lightweight point-voxel transformation (PVT) module behind the decoder to adaptively map sparse predictions into the dense voxel space.<n>Our architecture decouples feature and occupancy generation processes, allowing OPUS-V2 to adapt to arbitrary occupancy resolutions.
  arXiv  Detail & Related papers  (2026-08-29T10:45:34Z)
• Empowering Compact LLMs with Fusion of Layer-wise Exits for Recommendation [59.216721553953]
  Large language model-based recommender systems (LLM-RSs) have demonstrated remarkable capabilities, but are computationally unsustainable for many real-world applications.<n>We propose a discriminative framework that enhances compact LLMs while retaining scalable full-corpus ranking.<n>Experiments on three real-world datasets with Qwen 3 1.7B and Llama 3.2 3B show that FLEXRec achieves state-of-the-art accuracy while remaining highly efficient.
  arXiv  Detail & Related papers  (2026-08-18T03:14:28Z)
• LALE: Lightweight-Transformer Architecture for Land-Cover Estimation [4.554894288663752]
  We present LALE, an end-to-end remote sensing image segmentation architecture, that bifurcates its encoder by resolution.<n>On the large-scale ARAS400k remote-sensing segmentation benchmark, LALE establishes a strong efficiency-performance trade-off against CNN, transformer, and hybrid baselines.<n>Our smallest variant, (just 1.6M parameters), reaches within 2.6 F1 points of the best baseline (UPerNet) while using 4.5x fewer parameters, 7x less storage, 17x fewer GMACs, and delivering 1.8x higher throughput.
  arXiv  Detail & Related papers  (2026-06-01T11:18:49Z)
• Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders [123.58723804218151]
  Vision foundation models are bottlenecked by the quadratic cost of self-attention.<n>We present C-GSPN based on 2D spatial propagation propagation.
  arXiv  Detail & Related papers  (2026-05-30T14:29:43Z)
• EVCC: Enhanced Vision Transformer-ConvNeXt-CoAtNet Fusion for Classification [0.5394291557377919]
  Hybrid vision architectures combining Transformers and CNNs have significantly advanced image classification, but they usually do so at significant computational cost.<n>We introduce EVCC, a novel multi-branch architecture integrating the Vision Transformer, lightweight ConvNeXt, and CoAtNet.<n> Experiments across the CIFAR-100, Tobacco3482, CelebA, and Brain Cancer datasets demonstrate EVCC's superiority over powerful models.
  arXiv  Detail & Related papers  (2025-11-24T02:11:19Z)
• LIDAR: Lightweight Adaptive Cue-Aware Fusion Vision Mamba for Multimodal Segmentation of Structural Cracks [27.57718303520023]
  We propose a Lightweight Adaptive Cue-Aware Vision Mamba network.<n>It efficiently perceives and integrates morphological and textural cues from different modalities under multimodal crack scenarios.<n>Our method achieves 0.8204 in F1 and 0.8465 in mIoU with only 5.35M parameters.
  arXiv  Detail & Related papers  (2025-07-30T08:28:20Z)
• MetaOcc: Spatio-Temporal Fusion of Surround-View 4D Radar and Camera for 3D Occupancy Prediction with Dual Training Strategies [12.485905108032146]
  This paper introduces MetaOcc, a novel multi-modal framework for omni-oriented 3D occupancy prediction.<n>To address the limitations of directly applying encoders to sparse radar data, we propose a Radar Height Self-Attention module.<n>To reduce reliance on expensive point cloud, we propose a pseudo-label generation pipeline based on an open-set segmentor.
  arXiv  Detail & Related papers  (2025-01-26T03:51:56Z)
• MobileUNETR: A Lightweight End-To-End Hybrid Vision Transformer For Efficient Medical Image Segmentation [0.12499537119440242]
  Skin cancer segmentation poses a significant challenge in medical image analysis. MobileUNETR aims to overcome the performance constraints associated with both CNNs and Transformers. MobileUNETR achieves superior performance with 3 million parameters and a computational complexity of 1.3 GFLOP.
  arXiv  Detail & Related papers  (2024-09-04T20:23:37Z)
• Lightweight Full-Convolutional Siamese Tracker [4.903759699116597]
  This paper proposes a lightweight full-convolutional Siamese tracker called LightFC. LightFC employs a novel efficient cross-correlation module and a novel efficient rep-center head. Experiments show that LightFC achieves the optimal balance between performance, parameters, Flops and FPS.
  arXiv  Detail & Related papers  (2023-10-09T04:07:35Z)
• Efficient Decoder-free Object Detection with Transformers [75.00499377197475]
  Vision transformers (ViTs) are changing the landscape of object detection approaches. We propose a decoder-free fully transformer-based (DFFT) object detector. DFFT_SMALL achieves high efficiency in both training and inference stages.
  arXiv  Detail & Related papers  (2022-06-14T13:22:19Z)
• BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation [105.96557764248846]
  We introduce BEVFusion, a generic multi-task multi-sensor fusion framework. It unifies multi-modal features in the shared bird's-eye view representation space. It achieves 1.3% higher mAP and NDS on 3D object detection and 13.6% higher mIoU on BEV map segmentation, with 1.9x lower cost.
  arXiv  Detail & Related papers  (2022-05-26T17:59:35Z)
• LiDAR-based Panoptic Segmentation via Dynamic Shifting Network [56.71765153629892]
  LiDAR-based panoptic segmentation aims to parse both objects and scenes in a unified manner. We propose the Dynamic Shifting Network (DS-Net), which serves as an effective panoptic segmentation framework in the point cloud realm. Our proposed DS-Net achieves superior accuracies over current state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-11-24T08:44:46Z)

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.