ICMV is an annual International Conference on Machine Vision. Being held for the 11th time, it brings together researches, experts and practitioners from all over the world to discuss their latest research findings, challenges and lessons learned in the field of machine vision, pattern recognition, machine learning, data mining, and image analysis.
Smart Engines software engineers and developers have shown active involvement in ICMV through the history of the event, and this year is not an exception. Our team was excited to contribute our latest findings and share them with other scholars and practitioners, and also learn about the most innovative work being done in the field of machine learning.
This year, some of the our paper presentations comprised the research on the following topics:
- Fast Hamming distance computation on VLIW-architecture for 2D art recognition
- Methods of machine-readable zone recognition results post-processing
- Effective real-time augmentation of training data set for the neural networks learning
- Document image recognition algorithm based on similarity metric robust to projective distortions for mobile devices
- Viability of Viola-Jones method for the problem of image classification
- A method of projective transformations graph adjustment for planar object panorama stitching
- Modification of the Viola-Jones approach for the detection of the government seal stamp of the Russian Federation
- The use of FHT, its modification for practical applications and the structure of Hough image
- A method for spatially weighted image brightness normalization for face verification
- Dynamic programming approach to template-based OCR
- Experimental modeling the flow of character recognition results in video stream for document recognition
- Application of dynamic saliency maps to the video stream recognition systems with image quality assessment
- Automatic cropping of images under projective transformation
- 2D art recognition in uncontrolled conditions using one-shot learning
- Differentiation of Common Image Processing Algorithms
The most valuable papers will be published by SPIE, included in SPIE Digital Library, and added to Web of Science, Scopus and Ei Compendex citation Index. Our team would like to thank the organizers and all the participants for such a great conference, it’s been a wonderful experience to be a part of it again.