Online shopping for auto parts is often a difficult and frustrating task. If you want to order a new bezel gasket for the 1995 Ford Fusion, then you probably won't find the right one at all. Even if some components are tagged with “compatible” labels, these components are often non-compliant due to manufacturing inconsistencies. In addition, there are usually hundreds of brands to choose from in the auto parts market. Some dazzling car brands and models also need to be remembered.
Amazon has already launched retail services through AR technology on its iOS and Android platform apps, which allows consumers to project rendered images of furniture and electronic products into their homes. A new patent granted to Amazon later last month shows that it may expand AR into a brand new retail category.
Amazon began targeting this vertical market in January 2017 when the company announced that it had signed contracts with some of the country’s largest auto parts suppliers. It is worth noting that traditional automotive component retailers such as Advance Auto Parts and AutoZone have no support for AR functionality on their Web or App, giving Amazon a breakthrough in this area.
Amazon's recent AR patent license, entitled "Preview image generation for vehicle component installations", describes an image system that uses AR to allow users to preview car parts as if the car parts were actually installed in their cars.
According to the description in the patent, the software first obtains the image of the vehicle through the user's camera, identifies the brand, model, and year of the customer's vehicle (the application stores in its database an accurate measurement of the vehicle).
The computer vision technology in the software can then detect the vehicle based on the car battery, headlights, or brand logo. If the vehicle condition cannot be detected by computer vision, the user can also manually input the vehicle's identification number (VIN).
Once the vehicle information is confirmed, the software can identify a certain part of the vehicle (for example, the engine) and detect the installation location of the selected automobile parts.
Once the vehicle connection point is identified, the user can select the desired component from the online database and generate a 3D rendered preview. Take the engine's supercharger as an example, this software can display multiple brands of different specifications of the supercharging device, as shown in the picture, allowing the user to switch between different options.
The patent details the internal recommendation mechanism of this software, which is similar to the recommendation function on Amazon's retail website. This software can also provide other supplementary parts that match this part.
After selecting the required car parts, the platform integrates the image of the selected car part with the image of the vehicle. As shown in the following figure, the user can determine whether this part is suitable for their own car.
It is worth noting that the patent states that 3D rendering can determine whether a part fits into a vehicle within a specific "threshold degree of fit". In other words, even if accurate measurement data is stored in the software's database, it still cannot work well as we think. With visually predictable fit problems (such as small pieces of screws), this software also contains additional information such as user or manufacturer reviews, which provides more background information.
If this technology can make Amazon accurately provide compatible parts and components, the scope of consumer choice is no longer limited to expensive original parts and components, which can help the auto parts industry to gain more shares in the aftermarket.
Amazon has already launched retail services through AR technology on its iOS and Android platform apps, which allows consumers to project rendered images of furniture and electronic products into their homes. A new patent granted to Amazon later last month shows that it may expand AR into a brand new retail category.
Amazon began targeting this vertical market in January 2017 when the company announced that it had signed contracts with some of the country’s largest auto parts suppliers. It is worth noting that traditional automotive component retailers such as Advance Auto Parts and AutoZone have no support for AR functionality on their Web or App, giving Amazon a breakthrough in this area.
Amazon's recent AR patent license, entitled "Preview image generation for vehicle component installations", describes an image system that uses AR to allow users to preview car parts as if the car parts were actually installed in their cars.
Step 1: Identify Vehicles with Computer Vision
According to the description in the patent, the software first obtains the image of the vehicle through the user's camera, identifies the brand, model, and year of the customer's vehicle (the application stores in its database an accurate measurement of the vehicle).
The computer vision technology in the software can then detect the vehicle based on the car battery, headlights, or brand logo. If the vehicle condition cannot be detected by computer vision, the user can also manually input the vehicle's identification number (VIN).
Once the vehicle information is confirmed, the software can identify a certain part of the vehicle (for example, the engine) and detect the installation location of the selected automobile parts.
Step 2: Online purchase
Once the vehicle connection point is identified, the user can select the desired component from the online database and generate a 3D rendered preview. Take the engine's supercharger as an example, this software can display multiple brands of different specifications of the supercharging device, as shown in the picture, allowing the user to switch between different options.
The patent details the internal recommendation mechanism of this software, which is similar to the recommendation function on Amazon's retail website. This software can also provide other supplementary parts that match this part.
Step 3: Test the installation with AR
After selecting the required car parts, the platform integrates the image of the selected car part with the image of the vehicle. As shown in the following figure, the user can determine whether this part is suitable for their own car.
It is worth noting that the patent states that 3D rendering can determine whether a part fits into a vehicle within a specific "threshold degree of fit". In other words, even if accurate measurement data is stored in the software's database, it still cannot work well as we think. With visually predictable fit problems (such as small pieces of screws), this software also contains additional information such as user or manufacturer reviews, which provides more background information.
If this technology can make Amazon accurately provide compatible parts and components, the scope of consumer choice is no longer limited to expensive original parts and components, which can help the auto parts industry to gain more shares in the aftermarket.
评论
发表评论