Data tag classification

Data labeling classification includes the following categories:

Computer vision: This kind of data annotation mainly includes frame annotation, semantic segmentation, 3D point cloud annotation, key point annotation and line annotation. Pull frame marking is to mark the specified target object in image or video data with 2D frame, 3D frame and polygonal frame. Semantic segmentation is to mark the content and position in the image and segment it at the pixel level according to the attributes; Three-dimensional point cloud labeling is to label various types of specified objects in point cloud data images with three-dimensional frames; Key point labeling is to label the key points of human face and human skeleton. Line labeling supports Bezier curve and ordinary line segment labeling.

Please click to enter a picture description.

Natural language processing: including OCR transcription, text information extraction, NLU sentence summarization, part-of-speech tagging, emotional judgment and intention judgment.

Please click to enter a picture description.

Voice engineering: This kind of annotation mainly includes ASR voice transcription, voice emotion judgment, voiceprint recognition annotation and voice cutting.

Jinglianwen Technology is the head enterprise of AI basic data industry. An advanced data labeling platform and mature labeling, auditing and quality inspection mechanisms for customized data labeling services have been established. Through intelligent labeling platform products, AI can train the data industry, covering major AI algorithm fields such as computer vision, voice engineering and natural language understanding, supporting all types of labeling tools for images, videos, voices and text point clouds, and supporting the deployment of privatized software.

Please click to enter a picture description.

The labeling platform has a real-time quantitative visual management system, including data set management, project management, personnel management, supply chain management, etc., and has a perfect quality inspection process, which can effectively improve the efficiency of man-machine collaboration, expand production capacity, adjust the labeling scheme in time, do a good job in overdue risk management and control, and accurately control data quality problems; Establish a perfect personnel training and management system for the full-time bidding team, launch a complete set of AI industry personnel training solutions, and offer training programs such as theoretical courses, practical courses and graduation exams respectively. Through the combination of theory and practice, we will deliver high-quality data acquisition and labeling talents for the industry.

Jinglianwen Technology has been paying attention to various needs in artificial intelligence scenarios such as smart driving, smart home, public safety, smart city, smart medical care, smart finance, smart education, smart justice, etc., and providing the underlying technical support for AI technology. In the future, Jinglianwen Technology will continue to focus on high-quality, scene-based AI data services, and create a general data set, a data labeling platform and data management tools, and a customized service layout for AI data collection and labeling.