MARSAIL

MARS Artificial Intelligence Laboratory

MARSAIL (Motor AI Recognition Solution AI Lab)

We build and train AI that truly understands the entire automotive insurance ecosystem — from every scratch and component to every claim, repair note, and document. Our models are designed to be ethical, transparent, and fair, helping make car insurance smarter for everyone.

Led by Teerapong Panboonyuen

🚀 About MARS

MARS is a state-of-the-art deep learning model for car damage instance segmentation. It stands for:

M = Mask
A = Attention
R = Refinement
S = Sequential Quadtree Nodes

By leveraging sequential quadtree attention, MARS refines segmentation masks at an instance level, surpassing existing methods (Mask R-CNN, PointRend, Mask Transfiner) with notable maskAP gains.

Presented at ICIAP 2023, Udine, Italy 🇮🇹

🏁 About MARSAIL Lab

MARSAIL (Motor AI Recognition Solution Artificial Intelligence Laboratory) is dedicated to pioneering research at the intersection of computer vision, transformers, and automotive AI.

Our mission is to revolutionize the automotive insurance and repair industries through AI-driven automation, delivering breakthroughs in segmentation, localization, and decision intelligence.

MARSAIL Lab Logo

✨ Visual Results

MARS delivers superior segmentation accuracy:

Architecture
MARS Segmentation

🧠 MARSAIL Research Highlights (2023–2026)

At the heart of MARSAIL lies a commitment to pushing the boundaries of AI in automotive technology. All research initiatives are spearheaded by Dr. Kao (Teerapong Panboonyuen), whose leadership drives our mission to advance computer vision, document intelligence, and damage-centric transformer architectures. Below are our flagship contributions that have shaped the future of automotive AI.

🚗 ALBERT — Advanced Transformers for Comprehensive Automotive Damage Evaluation (Teacher Model)

ALBERT introduces a next-generation transformer backbone designed for holistic vehicle damage understanding. It delivers powerful contextual embeddings, multi-level localization cues, and robust attention representations that guide downstream segmentation and assessment models. Its architecture establishes ALBERT as a foundational “teacher model” for automotive AI systems. [arXiv]

@article{panboonyuen2025albert,
  title={ALBERT: Advanced Localization and Bidirectional Encoder Representations from Transformers for Automotive Damage Evaluation},
  author={Panboonyuen, Teerapong},
  journal={arXiv preprint arXiv:2506.10524},
  year={2025}
}

🧩 SLICK — Knowledge-Enhanced Vehicle Damage Segmentation (Student Model)

Built upon ALBERT’s distilled knowledge, SLICK specializes in fine-grained, automotive-specific segmentation. It employs selective localization, hierarchical calibration, and optimized inference pathways to deliver real-time segmentation accuracy suited for insurance workflows and repair automation. [arXiv]

@article{panboonyuen2025slick,
  title={SLICK: Selective Localization and Instance Calibration for Knowledge-Enhanced Car Damage Segmentation in Automotive Insurance},
  author={Panboonyuen, Teerapong},
  journal={arXiv preprint arXiv:2506.10528},
  year={2025}
}

🔍 DOTA — RAG-Enhanced OCR for Automotive Services

DOTA combines deformable transformer attention with retrieval-augmented generation to achieve human-level robustness in real-world document digitization. Designed specifically for automotive workflows, it excels in extracting structured information from invoices, claims, inspection sheets, and handwritten service notes. [arXiv]

@inproceedings{marsail2025dota,
  title={DOTA: Deformable Optimized Transformer Architecture for End-to-End Text Recognition with Retrieval-Augmented Generation},
  author={Panboonyuen, Teerapong as co-author and lead researcher},
  booktitle={2025 17th International Conference on Knowledge and Smart Technology (KST)},
  pages={301--306},
  year={2025},
  organization={IEEE}
}

🌌 MARS — Foundational Mask Attention Refinement Model

MARS introduced the groundbreaking Sequential Quadtree Attention mechanism, enabling more accurate mask refinement for instance-level damage detection. This work laid the foundation for MARSAIL’s broader research direction, demonstrating significant performance gains over Mask R-CNN, PointRend, and Mask Transfiner. [Springer]

@inproceedings{panboonyuen2023mars,
  title={Mars: Mask attention refinement with sequential quadtree nodes for car damage instance segmentation},
  author={Panboonyuen, Teerapong},
  booktitle={International Conference on Image Analysis and Processing},
  pages={28--38},
  year={2023},
  organization={Springer}
}

⚡ Quick Start

git clone https://github.com/kaopanboonyuen/MARS.git
cd MARS
python3 -m venv mars-env
source mars-env/bin/activate
pip install -r requirements.txt
  
GitHub Repository

📂 Datasets

✍️ Citation

@inproceedings{panboonyuen2023mars,
  title={MARS: Mask Attention Refinement with Sequential Quadtree Nodes},
  author={Panboonyuen, Teerapong},
  booktitle={ICIAP},
  year={2023},
  publisher={Springer}
}
    

🧠 Want to Know More About MARSAIL?

Explore how our lab is shaping the future of automotive AI, blending deep vision models with smart insurance tech. Get the full story, insights, and breakthroughs from the MARSAIL team in our feature blog post.

Read the Full Blog

🎨 MARSAIL Official Logo

The official MARSAIL logo is available for use in academic papers, research posters, presentations, conference submissions, and other professional materials.

Please use the logo below when representing the laboratory in publications or collaborative work.

MARSAIL Lab Logo
⬇️ Download Logo (PNG)

🤝 Our Trusted Sponsors & Partners

We are deeply grateful for the invaluable support of our sponsors, partners, and collaborators. Their investment and belief in our mission empower MARSAIL to push boundaries in automotive AI.

🌟 Strategic Investor

Thaivivat Insurance Logo

A special thank you to Thaivivat Insurance Public Company Limited (TVI) for their forward-thinking investment in AI-driven automotive solutions.

MARSAIL Logo

MARSAIL (MARS AI Lab)

Chulalongkorn University Logo

Chulalongkorn University

PBY.LAB Logo

PBY Artificial Intelligence Laboratory

🌟 Fun Facts About MARSAIL

🇺🇸 Fun Facts — English Version

MARSAIL is the dedicated Artificial Intelligence Laboratory under MARS (Motor AI Recognition Solution) . While MARS was founded earlier as a tech company, the AI-focused MARSAIL was later established and shaped by Dr. Teerapong Panboonyuen (Dr. Kao) , who leads the AI Research & Development division.


After completing his Ph.D. at Chulalongkorn University , Dr. Kao joined MARS in January 2022 to spearhead the next era of AI innovation. Upon reviewing the existing “first-generation AI system” from 2021, he made a decisive move — he removed the entire legacy system and rebuilt everything from the ground up.


This total reboot resulted in the new MARS model (Mask Attention Refinement with Sequential Quadtree Nodes), inspired partly by the company name. It became MARSAIL’s first official research publication and was presented on the international stage at ICIAP 2023 in Udine, Italy .


From 2022–2026, under Dr. Kao’s leadership, MARSAIL evolved into a next-generation AI research lab, producing advanced models including the flagship transformer ALBERT . Today, MARSAIL powers innovation across car insurance, damage analytics, intelligent document understanding, and AI assistance for vehicle damage assessment.

🇹🇭 เกร็ดน่ารู้ — เวอร์ชันภาษาไทย

MARSAIL คือห้องปฏิบัติการปัญญาประดิษฐ์ของบริษัท MARS – Motor AI Recognition Solution (บริษัท มอเตอร์ เอไอ เรคอกนิชั่น โซลูชั่น จำกัด) โดยบริษัท MARS มีการก่อตั้งมาก่อน แต่ห้องแลป MARSAIL ถูกสร้างและพัฒนาโดย ดร.ธีรพงศ์ ปานบุญยืน (ดร.เก้า) ในฐานะ Head of AI Research เพื่อผลักดันการวิจัย AI ให้เติบโตระดับองค์กร


หลังจบ ปริญญาเอกจากจุฬาลงกรณ์มหาวิทยาลัย ดร.เก้าเริ่มเข้ามาทำงานที่ MARS ในช่วง มกราคม 2022 เพื่อสานต่องานพัฒนา AI จากทีมรุ่นแรกในปี 2021 แต่แทนที่จะปรับแต่งแบบเดิม เขากลับตัดสินใจครั้งใหญ่ — ลบระบบ AI เดิมทั้งหมด และสร้างระบบใหม่ตั้งแต่ศูนย์ เพื่อให้ทันสมัย แข็งแรง และรองรับงานระดับอุตสาหกรรม


โมเดลใหม่ที่เกิดจากการรีบูตนี้คือ MARS (Mask Attention Refinement with Sequential Quadtree Nodes) ซึ่งสอดคล้องกับชื่อบริษัท และได้เป็นงานวิจัยลำดับแรกของห้องแลป MARSAIL ที่ถูกนำเสนอในเวทีนานาชาติ ICIAP 2023 ณ ประเทศอิตาลี


ตั้งแต่ปี 2022–2026 ดร.เก้า ได้พัฒนา MARSAIL ให้เติบโตเป็นห้องวิจัย AI ระดับแนวหน้า สร้างโมเดลเด่น เช่น ALBERT ที่เป็นหัวใจของงานด้านประกันภัยรถยนต์ การประเมินรอยความเสียหาย การทำความเข้าใจเอกสารอัจฉริยะ และ AI assistance สำหรับงานประเมินราคาความเสียหายของรถยนต์ ผลักดันอุตสาหกรรมยานยนต์ไทยให้ล้ำหน้าด้วยพลังของ AI

🤖 MARSAIL AI
Welcome to MARSAIL AI! Scroll and I’ll tell you more.