Seeing the Signs: A Survey of Edge-Deployable OCR Models for Billboard Visibility Analysis

2025/07/15 by Maciej Szankin, Szankin, Maciej, Vidhyananth Venkatasamy +3
Computer Science · Social Sciences · #Advanced Image and Video Retrieval Techniques #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Geographic Information Systems Studies #Image Retrieval and Classification Techniques

paper · pdf · doi:10.48550/arxiv.2507.11730

openalex publication_date 2025/07/15 · openalex created_date 2025/10/14 · openalex updated_date 2026/07/28

Abstract

Outdoor advertisements remain a critical medium for modern marketing, yet accurately verifying billboard text visibility under real-world conditions is still challenging. Traditional Optical Character Recognition (OCR) pipelines excel at cropped text recognition but often struggle with complex outdoor scenes, varying fonts, and weather-induced visual noise. Recently, multimodal Vision-Language Models (VLMs) have emerged as promising alternatives, offering end-to-end scene understanding with no explicit detection step. This work systematically benchmarks representative VLMs - including Qwen 2.5 VL 3B, InternVL3, and SmolVLM2 - against a compact CNN-based OCR baseline (PaddleOCRv4) across two public datasets (ICDAR 2015 and SVT), augmented with synthetic weather distortions to simulate realistic degradation. Our results reveal that while selected VLMs excel at holistic scene reasoning, lightweight CNN pipelines still achieve competitive accuracy for cropped text at a fraction of the computational cost-an important consideration for edge deployment. To foster future research, we release our weather-augmented benchmark and evaluation code publicly.

Citations

• Gemma 3 Technical Report
• SVTRv2: CTC Beats Encoder-Decoder Models in Scene Text Recognition
• Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
• InstaTune: Instantaneous Neural Architecture Search During Fine-Tuning
• Scene Text Recognition with Permuted Autoregressive Sequence Models
• DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale
• Unitail: Detecting, Reading, and Matching in Retail Scene
• PP-OCR: A Practical Ultra Lightweight OCR System
• ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network
• Decoupled Attention Network for Text Recognition
• Scan, Attend and Read: End-to-End Handwritten Paragraph Recognition with MDLSTM Attention
• An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
• Reading Text in the Wild with Convolutional Neural Networks

Related

• Text Detection and Recognition in the Wild: A Review
• HunyuanOCR Technical Report
• VLMs-in-the-Wild: Bridging the Gap Between Academic Benchmarks and Enterprise Reality
• OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning
• Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments