AI seeks the next growth point and implements the three-dimensional world of machines

In order to realize the machine's depth perception and understanding of the three-dimensional world, TuoVision has given us a new perspective. To understand the three-dimensional world from two-dimensional images, there must be irreversible information loss, which brings the machine perception and understanding. Passover obstacles. Then the development of the 3D vision industry will inevitably lead to an outbreak.

Currently, low-cost 3D imaging solutions include three main ways of flying time (TOF), structured light, and stereo vision. For the user, the difference lies mainly in imaging accuracy, speed, resolution, frame rate and dependence on the environment. At present, most of the 3D vision products in the market are concentrated on the 3D imaging sensor itself, but a single sensor can only obtain scene information at a specific viewing angle. How to integrate the data obtained by the three-dimensional imaging sensors in different positions and different poses to obtain a true and complete three-dimensional model of the entire scene is a very difficult problem.

AI seeks the next growth point and implements the three-dimensional world of machines

The original intention of “Tuo Vision” was to achieve a better perception, reconstruction and understanding of the machine in the three-dimensional world. Its products can replicate the three-dimensional objective world into virtual space, thus establishing a three-dimensional data service ecosystem and providing three-dimensional data support for different industry applications. In the field of virtual reality, the extension vision product solves the problem that traditional panoramic photos can only be viewed and cannot be measured; in the field of augmented reality, the extension vision product bridges the gap between virtual reality and reality, and provides three-dimensional data support for augmented reality. At present, its products are mainly used in industries such as virtual home improvement, real estate measurement, law enforcement forensics and intelligent robots.

You can understand the extension vision products from the perspective of technology research and development, technology landing, and industry application.

3D vision technology barriers are relatively high. Extension vision has developed structured light 3D imaging and TOF imaging technology in 3D imaging technology. It has accumulated 3D object reconstruction, 3D scene reconstruction, and simultaneous positioning and mapping (SLAM) technology in point cloud processing. In the aspect of detection and tracking identification, 3D face and object detection, recognition, tracking and human skeleton extraction technology have been developed. In the rendering display, there are massive point cloud meshing, high-precision texture mapping and high-resolution panorama generation technology.

In terms of technology, the company has developed the tModel indoor 3D modeling system, tScan high-precision 3D object scanning system and tSense high frame rate depth camera.

The accuracy of tModel 3D modeling is centimeter. The main application direction is real estate and home improvement. It is difficult to solve the problem of real estate in the field, high time cost, and the painful effect of the existing virtual reality home improvement products. In addition, tModels can be used in public safety areas such as anti-terrorism drills and crime scene reconstruction.

tScan's high-precision 3D object scanning system with imaging accuracy better than 0.1 mm and modular customization. Application scenarios include 3D face scanning for medical aesthetics, component modeling for industrial assembly, and 3D body scanning for custom clothing and footwear.

The tSense 3D camera has an imaging rate of 60 frames per second. It can realize real-time 3D imaging of moving objects. The application scenarios include 3D motion capture, gesture recognition and behavior recognition. It has broad application prospects in the fields of human-computer interaction and video entertainment.

The company's business model is divided into short-term and long-term goals. Among them, short-term goals are in the form of “products + services”, which are sold and rented through equipment to realize three-dimensional modeling data services. With the growth of services and the accumulation of data, long-term goals will be transformed. In the form of “platform + data”, it is to build a large-scale 3D model data production, storage and trading platform, and combined with the introduction of AI technology, based on the platform to provide big data analysis and derivative services.

3D vision is an emerging industry, and its mature products are not many. The typical competing product of TuoVision is Matterport of the United States. Founded in 2011, Matterport has completed the D round of financing with a total financing of $66 million. The products mainly include Matterport 3D cameras, which are relatively single.

CEO Lu Min told that the world of human existence is three-dimensional. If you understand the three-dimensional world from two-dimensional images, there must be irreversible information loss, which brings insurmountable obstacles to machine perception and understanding. Three-dimensional vision focuses on restoring the essence of the three-dimensional world and realizing the understanding of the real world in the three-dimensional digital space. Therefore, it is an important development direction in the field of artificial intelligence and will become the next new market growth point.

In the recently released iPhone X, Apple has embedded a structured light depth camera into the phone to implement face unlocking and face payment through face ID. In addition, international IT giants such as Microsoft and Intel have a long-term layout in 3D vision, and have developed their own 3D imaging sensors. Obviously, with the strong entry of international IT giants, the 3D vision industry will explode in the near future and become a brand new blue ocean market. This is also an opportunity for many startups.

Dome Lens

The dome Optical Lens is a hemisphere. The thickness of the two optical surfaces is equal, forming a naturally strong shape, which becomes stronger under pressure. This makes Dome Lens ideal for underwater environments and applications such as camera dome ports and diving windows.BK7 or K9L domes hemispherical lenses are mainly used for meteorological applications. BK7 provides excellent transmittance in the 300nm to 2μm range. BK7 is a relatively hard material with excellent chemical durability.UV Fused Silica Domes: For applications operating in the deeper UV range, Worldhawk offers a series of UV Fused Silica Domes. Fused silica domes are commonly used for extremely high pressure underwater applications.Optical sapphire dome: For infrared applications, we can provide dome Spherical lenses. Sapphire is an extremely hard material with a transmittance of over 80% in the wavelength range of 2-5 microns. Like fused silica, sapphire can withstand extreme pressure, making it the perfect material for underwater camera and missile fairing applications.

Dome Cover Lens,Bk7 Optical Glass Dome,Uv Fused Silica Dome,Customized Optical Hemisphere

Zoolied Inc. , https://www.zoolied.com