A large scale human-labeled dataset plays an important role in creating high quality deep learning models. In this paper we present text annotation for Open Images V5 dataset. To our knowledge it is the largest among publicly available manually created text annotations. Having this annotation we trained a simple Mask-RCNN-based network, referred as Yet Another Mask Text Spotter (YAMTS), which achieves competitive performance or even outperforms current state-of-the-art approaches in some cases on ICDAR 2013, ICDAR 2015 and Total-Text datasets. Code for text spotting model available online at: https://github.com/openvinotoolkit/training_extensions. The model can be exported to OpenVINO™-format and run on Intel® CPUs.
The main stages of development of the process of production of armor steel in the Magnitogorsk metallurgical works are described from the beginning of the Second World War to date.
The plans for increasing oil recovery and expanding the present network of gas pipelines in the Twenty-First Century make it extremely important to also speed up the construction of new gas pipelines and the repair of those already in existence. In connection with this, the Russian market for large-diameter pipes will reach roughly 1 million tons a year. That in turn makes it essential that a domestic complex for the production of large-diameter pipes be created as soon as possible. The development of such a complex would free "Gazprom" from its dependence on imports and, thus, would also improve national security from the standpoint of energy supplies.The plan is for natural gas to be recovered in reserves already being exploited on the continental shelf. The reserves are located in the northern seas and on the territory of Western Siberia. The critical factor in this undertaking is ensuring that the metal of the line pipes is of high quality and that the pipes are durable and reliable, since the harsh climate of the Far North imposes stringent requirements on the service properties of steel structures in general and gasline pipe in particular. The cost of emergency-rescue operations under the severe northern weather conditions is too high to risk accidents.To solve this problem, the Nizhniy Tagil Metallurgical Combine (NTMK) has begun organizing the production of high-strength cold-resistant steel of type Kh70-Kh80 and large-diameter (1020-1420 mm) pipes.The Central Scientific Research Institute of Ferrous Metallurgy (TsNIIchermet) used the results of an analysis of domestic and foreign practices to develop an integrated technology for the production of 10-40-mm skelp of strength category K60-K65 (Kh70-Khg0 under API 5L), high-strength (Yd up to 1000 MPa) finished plate 5-50 mm thick, slabs (Yd = 50(0800 MPa) 60-200 mm thick, and one-seam 1020-1420-mm-diam. pipes of strength category Kh70-Kh80 under API 5L. The most progressive methods will be used in each stage of the production process. The institute also examined the practicality of various technologies that might be used by the NTMK (which produces steel that has low concentrations of impurity atoms of nonferrous metals and nonmetallic inclusions and that can be used to make rolled products with little striation) to make pipes which will be more resistant than foreign-made pipes to structural degradation during long service, stress corrosion cracking, and brittle fracture at minus temperatures.The converter shop at the combine operates by a duplex process on vanadium pig iron smelted from ores of the Tagilo-Kushvinsk and Kachkanar deposits. These ores contain vanadium and titanium, making the shop unique in world metallurgy in being able to use such pig iron.The technology of the duplex process provides for the conversion of vanadium pig iron. The first stage of the process involves the production of commercial-grade vanadium slag (the content ofV205 = 16-20%) and a clean metallic semi finished product. In the second stage, steel is made from the semifinish...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.