Bibliography · 2 papers
Publications
Peer-reviewed work in OCR, document AI, and time series forecasting.
- 02
Revisiting Long-Term Time Series Forecasting: An Empirical Perspective on Recent Approaches
This paper revisits long-term time series forecasting by re-evaluating five representative Transformer-based and linear-based models on five real-world datasets under a unified protocol. The study compares accuracy, stability, and computational efficiency, then highlights practical strengths, limitations, and open research challenges for future LSTF work.
- 01
VNProductKIE: A Dataset and Three-Stage Pipeline for Key Product Information Recognition on Vietnamese Packaging Labels
Food waste poses serious environmental and economic concerns, often worsened by the lack of accessible product information. Automated extraction from packaging labels offers a promising solution, yet existing datasets fall short in representing the linguistic and visual diversity found in Vietnamese markets. This paper introduces VNProductKIE, a new dataset of high-resolution images capturing Vietnamese food and beverage packaging. It features both English and Vietnamese text, diacritic-rich scripts, local date formats, and real-world distortions such as blur, curvature, and clutter. To extract structured information, this paper proposes a three-stage pipeline including: (1) a YOLOv11-based detector for locating key regions (e.g., product name, weight, brand, expiration date), (2) a word-level detector for segmenting individual words, and (3) a VietOCR-based recognizer for transcription. The final output is structured into complete product metadata. This pipeline experimented on VNProductKIE achieved a recognition accuracy of 98.85%, highlighting the effectiveness of the proposed approach.