Squeeze-Aware-YOLO: Early Channel Recalibration and Multi-Scale Fusion for Real-Time Traffic Sign Detection

Mizbaul Haque Maruf; Syed Samin Sadaf; Syed Tamzid Bakht; Md. Hasanul Kabir; Sabbir Ahmed

cs.CVOct 11, 2026

Squeeze-Aware-YOLO: Early Channel Recalibration and Multi-Scale Fusion for Real-Time Traffic Sign Detection

Authors: Mizbaul Haque Maruf, Syed Samin Sadaf, Syed Tamzid Bakht, Md. Hasanul Kabir, Sabbir Ahmed

Organizations: Dept. of Computer Science and Engineering, Islamic University of Technology, Boardbazar, Gazipur, Bangladesh

Abstract

Traffic sign detection plays a vital role in intelligent transportation systems, supporting the safe operation of advanced driver assistance systems and autonomous vehicles. Despite significant progress in real-time object detection, accurately identifying traffic signs under challenging conditions remains difficult. Variations in lighting, adverse weather, motion blur, and partial occlusion often reduce detection performance, especially for small or visually degraded signs where important spatial information can be lost. To address these challenges, we propose Squeeze-Aware-YOLO, a lightweight traffic sign detection framework that introduces a multi-branch bottleneck module combining convolutional kernels of different sizes with a Squeeze-and-Excitation mechanism to improve feature extraction and adaptive channel weighting. In addition, we integrate the Multi-Branch Auxiliary Feature Pyramid Network and Re-parameterized Heterogeneous Multi-Scale modules to strengthen multi-scale feature representation and feature fusion while preserving real-time inference efficiency for practical deployment. The proposed approach is evaluated on the Bangladeshi Traffic Sign Detection Dataset, the CSUST Chinese Traffic Sign Detection Benchmark 2021, and the German Traffic Sign Detection Benchmark, achieving [email protected] scores of 74.7%, 79.4%, and 84.0%, respectively, while consistently outperforming recent state-of-the-art YOLO-based traffic sign detectors. These findings demonstrate that combining multi-scale receptive fields, channel attention, and enhanced feature fusion enables more accurate and robust traffic sign detection while maintaining the computational efficiency required for real-world deployment.

Figures & tables

Fig. 1: Overview of the proposed Squeeze-Aware-YOLO architecture. The model consists of an SE-enhanced input stage with a custom bottleneck block, a hierarchical backbone built using RepHMS modules with multi-scale receptive fields, and a multi-branch neck incorporating Superficial Assisted Fusion (SAF) and Advanced Assisted Fusion (AAF) for effective feature aggregation.

Fig. 2: Simplified view of the SE-enhanced bottleneck block, showing the dual-path structure with sequential bottleneck processing and a compressed bypass path.

Panel of Figure 3 — Fig. 3: Two bottleneck variants used in the proposed architecture. (a) Bottleneck Type-2 employs a $5\times 5$ convolution, whereas (b) Bottleneck Type-1 uses a $3\times 3$ convolution.

Fig. 4: Detailed SE-enhanced bottleneck block, showing the dual-path structure: the main path processes through three sequential bottlenecks (Type-1, Type-2, Type-1) with varying kernels, while the bypass path provides direct gradient flow.

Model	BTSD	CCTSDB2021	GTSDB
YOLOv8	0.663	0.794	0.809
YOLOv10	0.679	0.792	0.826
YOLOv11	0.668	0.804	0.836
MHAF-YOLO	0.685	0.769	0.816
Squeeze-Aware-YOLO	0.747	0.794	0.840

TABLE I: [email protected] Comparison Across Datasets

Class	YOLOv8	YOLOv10	YOLOv11	MHAF	Ours
Ped. Crossing	0.760	0.735	0.425	0.675	0.730
SL40	0.845	0.775	0.722	0.759	0.850
SL80	0.563	0.729	0.725	0.709	0.705
SL60	0.622	0.671	0.737	0.894	0.950
No O/T	0.964	0.915	0.947	0.906	0.958
Bend (R)	0.794	0.821	0.793	0.687	0.894

TABLE II: Class-Wise [email protected] on the Bangladeshi Traffic Sign Detection Dataset

Model	Prohibitory	Mandatory	Warning	Overall
YOLOv8	0.798	0.737	0.847	0.794
YOLOv10	0.794	0.728	0.855	0.792
YOLOv11	0.810	0.739	0.864	0.804
MHAF-YOLO	0.782	0.687	0.838	0.769
Ours	0.810	0.712	0.858	0.794

TABLE III: Class-Wise [email protected] on CCTSDB2021

Model	Prohibitory	Mandatory	Danger	Others	Overall
YOLOv8	0.799	0.711	0.949	0.636	0.809
YOLOv10	0.919	0.726	0.906	0.755	0.826
YOLOv11	0.907	0.733	0.878	0.825	0.836
MHAF-YOLO	0.900	0.730	0.866	0.767	0.816
Ours	0.910	0.760	0.865	0.825	0.840

TABLE IV: Class-Wise [email protected] on GTSDB

Fig. 5: Qualitative detection comparison across three traffic-sign datasets. Rows correspond to the BTSD, CCTSDB2021, and GTSDB datasets, and columns show results from YOLOv8, YOLOv10, YOLOv11, MHAF-YOLO, and our method.

Baseline band	Classes	Mean $\Delta$ mAP	Improved
Hard ( $<0.60$ )	4	$+0.359$	4/4
Moderate ( $\geq 0.60$ , $<0.90$ )	7	$+0.004$	3/7
Easy ( $\geq 0.90$ )	4	$-0.310$	0/4

TABLE V: Per-Class [email protected] Change (

\Delta

) on BTSD Relative to the Best of YOLOv8, YOLOv10, and YOLOv11, Grouped by That Baseline’s Difficulty

Components			[email protected]
MHAF Baseline	TLB	SE	BTSD	CCTSDB2021	GTSDB
✓	✗	✗	0.685	0.769	0.816
✓	✓	✗	0.722	0.783	0.832
✓	✓	✓	0.747	0.794	0.840
a TLB denotes the Triple Layer Bottleneck module, and SE denotes the Squeeze-and-Excitation module. Values in the final row correspond to the complete proposed configuration (MHAF + TLB + SE).

TABLE VI: Ablation Study Showing [email protected] for Different Module Configurations on the BTSD, CCTSDB2021, and GTSDB Datasets

Fig. 6: Computational complexity of the evaluated models on CCTSDB2021 and BTSD: (a) GFLOPs for model and (b) parameter count (millions).

Fig. 7: Per-stage runtime decomposition (preprocessing, inference, postprocessing) on (a) CCTSDB2021 and (b) BTSD. Labels above each bar report the total end-to-end latency.

Explore similar work

Hierarchically Decoupled Mixture-of-Experts for Robust Traffic Sign Recognition in Complex Driving Scenarios
Jun 1, 2026Mingxiao Wang, Xiaozhen Qu, Bolin Gao +2Sparse Mixture-of-Experts Autonomous Driving Perception
School of Automotive and Traﬀic Engineering, Liaoning University of Technology, Jinzhou, Liaoning 121001, China. · State Key Laboratory of Intelligent Green Vehicles and Mobility, School of Vehicle and Mobility, Tsinghua University, Beijing 100084, China.
SMDDFNet: State-space Modeling and Dynamic Dual Fusion Network for Traffic Sign Detection
May 2, 2025TianYi Yu, DaJian Zhong, Lilin WangSmall Object Detection Multi-Scale Feature Fusion
Collage of Infomation Engineering, Shanghai Maritime University, Shanghai, 201306, China
RIECNN: Real-Time Image Enhanced CNN for Traffic Sign Recognition
Oct 10, 2026Reem Abdel-Salam, Rana Mostafa, Ahmed H. Abdel-GawadConvolutional Neural Networks Autonomous Driving Perception
Computer Engineering Department, Cairo University, Giza, Egypt

Jun 1, 2026J/K move · Enter open · S save

Hierarchically Decoupled Mixture-of-Experts for Robust Traffic Sign Recognition in Complex Driving Scenarios

Mingxiao Wang, Xiaozhen Qu, Bolin Gao, Tong Wang+1

School of Automotive and Traﬀic Engineering, Liaoning University of Technology, Jinzhou, Liaoning 121001, China. · State Key Laboratory of Intelligent Green Vehicles and Mobility, School of Vehicle and Mobility, Tsinghua University, Beijing 100084, China.

Sparse Mixture-of-Experts Autonomous Driving Perception Object Detection Mixture-of-Experts Models

Open paper AskPDF arXiv

Traffic sign detection is a fundamental component of environmental perception in autonomous driving and intelligent transportation systems. However, most existing detectors rely on static inference with globally shared parameters, limiting their ability to adapt to diverse and unstructured traffic scenarios. As a result, a single static model often struggles to simultaneously handle both clear near-range samples and challenging conditions such as distant small targets or adverse weather environments. To address this limitation, we propose CBDES MoE TSR, a hierarchically decoupled heterogeneous mixture-of-experts(MoE) framework for traffic sign recognition. The proposed framework departs from the conventional globally shared parameter paradigm by introducing a heterogeneous You Only Look Once (YOLO) expert pool together with a lightweight gating network, enabling an image-level dynamic routing mechanism. Based on the semantic characteristics of the input image, the gating module selectively activates the most suitable expert model from the expert pool, enabling a shift from fixed parameter fitting to on-demand dynamic representation. This design enhances feature extraction capability for specific scenarios while maintaining controlled inference overhead. Experimental results demonstrate that the proposed method achieves a remarkable balance between detection accuracy and efficiency on the composite traffic sign dataset. Specifically, our method attains an mAP50-95 of 76.8%, yielding a 2.3% improvement over the baseline method (74.5%) while simultaneously reducing computational overhead by approximately 39.4%. These findings robustly validate the effectiveness of the proposed approach.

Abstract

Explore similar work

Hierarchically Decoupled Mixture-of-Experts for Robust Traffic Sign Recognition in Complex Driving Scenarios

SMDDFNet: State-space Modeling and Dynamic Dual Fusion Network for Traffic Sign Detection

RIECNN: Real-Time Image Enhanced CNN for Traffic Sign Recognition