Spring meeting of Pattern Recognition and Computer Vision Colloquium 2016 in Prague: how it was
Good evening, habrasoobschestvo!
I already wrote about the summer school of computer vision, which took place last summer at the Czech Technical University in Prague in an article . At the end of March, I was lucky to take part in another one-day event of this kind, organized by the same group The Center for Machine Perception from the CTU in Prague. This time it was a series of 6 reports in breaks for lunch and a coffee break. The main topics of the event were computer vision and machine learning algorithms. Judging by the large number of reports on deep learning, this trend is becoming a very popular trend in solving computer vision problems. For details, I ask under the cat. Caution traffic!
The event was held in a very beautiful historic building of the Faculty of Electrical Engineering in the center of Prague.
As lecturers in summer school, there were invited scholars from various universities from around the world who were involved in computer vision research. Among them was also a doctoral student from the CTU.
Half of the presentations were devoted to machine learning, including graphic models, in-depth training and feature maps.
Antonin Sulc spoke about an interesting direction in computer vision - light fields and its application in practice. To begin with, he introduced the technology itself, which arose through the development of compact light-field cameras Lytro Illum. Such cameras make it possible to obtain, in addition to color information, information about the angle of incidence of the rays in each pixel. The simplest use of such cameras is to change the focus of the camera from the background to objects in the foreground and vice versa. The light-field technology is used in structure-from motion, it allows you to split an image into several layers (for example, to select reflective and transparent surfaces on the image) and to effectively estimate the depth (build a depth map).
')
Doctoral student Giorgos Tolias from CTU talked about his development of a tag descriptor that allows you to search for images using a hand-drawn sketch as a request.
The most interesting in my opinion was the report of Yannis Avrizis, who spoke about the methods of approximate search when solving clustering problems in a multidimensional space. He described the problem of searching in multidimensional space, showed in what tasks of computer vision this problem is solved (image search, frame position estimation, Bag of words and clustering). Various approximate search methods such as binary codes, quantization (vector, product quantization) and clustering (k-means and its various modifications, Gaussian mixtures and ranked retrieval and inverted-quantized k-means) were also presented. All methods were presented with a detailed mathematical explanation and shown graphically in graphical form.
You can view the whole program and download the presentations at: cmp.felk.cvut.cz/cmp/events/colloquium-2016.03.31 .
After the end of the main part, those wishing to take a short excursion to the offices of the CMP.
At the beginning, we were shown an online search engine for similar images for a given image fragment. The user can select an object of interest (for example, the dome of the cathedral) as a frame in the image and in return receives all relevant images that can contain this object from the same perspective, from different points of view, at different scales and even more detailed.
Here is a link to an article about the project for those who wish to learn more about the project: cmp.felk.cvut.cz/~chum/papers/mikulik_sisap13.pdf .
As a result, the day turned out to be intense and informative. I hope that in the future we will be able to attend other similar events. This event - Pattern Recognition and Computer Vision Colloquium - is held every six months, the last time it was held in November 2015. Such events allow me to get acquainted with experts and learn about new trends and technologies in the field of computer vision.
I wish everyone good luck and wait for new reviews!
I already wrote about the summer school of computer vision, which took place last summer at the Czech Technical University in Prague in an article . At the end of March, I was lucky to take part in another one-day event of this kind, organized by the same group The Center for Machine Perception from the CTU in Prague. This time it was a series of 6 reports in breaks for lunch and a coffee break. The main topics of the event were computer vision and machine learning algorithms. Judging by the large number of reports on deep learning, this trend is becoming a very popular trend in solving computer vision problems. For details, I ask under the cat. Caution traffic!
The event was held in a very beautiful historic building of the Faculty of Electrical Engineering in the center of Prague.
As lecturers in summer school, there were invited scholars from various universities from around the world who were involved in computer vision research. Among them was also a doctoral student from the CTU.
Half of the presentations were devoted to machine learning, including graphic models, in-depth training and feature maps.
Antonin Sulc spoke about an interesting direction in computer vision - light fields and its application in practice. To begin with, he introduced the technology itself, which arose through the development of compact light-field cameras Lytro Illum. Such cameras make it possible to obtain, in addition to color information, information about the angle of incidence of the rays in each pixel. The simplest use of such cameras is to change the focus of the camera from the background to objects in the foreground and vice versa. The light-field technology is used in structure-from motion, it allows you to split an image into several layers (for example, to select reflective and transparent surfaces on the image) and to effectively estimate the depth (build a depth map).
')
Doctoral student Giorgos Tolias from CTU talked about his development of a tag descriptor that allows you to search for images using a hand-drawn sketch as a request.
The most interesting in my opinion was the report of Yannis Avrizis, who spoke about the methods of approximate search when solving clustering problems in a multidimensional space. He described the problem of searching in multidimensional space, showed in what tasks of computer vision this problem is solved (image search, frame position estimation, Bag of words and clustering). Various approximate search methods such as binary codes, quantization (vector, product quantization) and clustering (k-means and its various modifications, Gaussian mixtures and ranked retrieval and inverted-quantized k-means) were also presented. All methods were presented with a detailed mathematical explanation and shown graphically in graphical form.
You can view the whole program and download the presentations at: cmp.felk.cvut.cz/cmp/events/colloquium-2016.03.31 .
After the end of the main part, those wishing to take a short excursion to the offices of the CMP.
At the beginning, we were shown an online search engine for similar images for a given image fragment. The user can select an object of interest (for example, the dome of the cathedral) as a frame in the image and in return receives all relevant images that can contain this object from the same perspective, from different points of view, at different scales and even more detailed.
Here is a link to an article about the project for those who wish to learn more about the project: cmp.felk.cvut.cz/~chum/papers/mikulik_sisap13.pdf .
As a result, the day turned out to be intense and informative. I hope that in the future we will be able to attend other similar events. This event - Pattern Recognition and Computer Vision Colloquium - is held every six months, the last time it was held in November 2015. Such events allow me to get acquainted with experts and learn about new trends and technologies in the field of computer vision.
I wish everyone good luck and wait for new reviews!
Source: https://habr.com/ru/post/281791/
All Articles