Called the IMX500 and IMX501, Sony says the upcoming intelligent vision sensors can be paired with cloud services, and having built-in AI that will aid “reduce data transmission latency, minimize any privacy concerns, and reduce power consumption and communication costs”.
Both sensors are backside-illuminated 1/2.3-inch sensors that have resolutions of approximately 12.3 megapixels and 4K/60fps video capabilities, and they’re expected to be priced at $93 and $187 respectively.
The secret behind the sensors is a new “stacked” design that combines a pixel chip with a logic chip, the latter of which has AI image analysis and processing capabilities. After light is captured by the pixel chip, it’s processed immediately by the AI on the logic chip hence the sensor being capable of outputting both image and metadata together or only the metadata if the user so pleases.
According to Sony; “The sensor outputs metadata instead of image information, making for reduced data volume and minimizing any privacy concerns. Moreover, the AI capability makes it possible to deliver diverse functionality for versatile applications, such as real-time object tracking with high-speed AI processing amongst many other things meaning a single type of camera can be used with versatility across different locations, circumstances, times, or purposes. When installed at the entrance to the facility it can be used to count the number of visitors entering the facility; when installed on the shelf of a store it can be used to detect stock shortages; when on the ceiling it can be used for heat mapping store visitors and the likes.”
Currently, the sensors are geared toward industrial and retail applications but with future developments, it’s possible the sensor could have practical applications in still and maybe motion photography.
Currently, there’s no indication that Sony will introduce AI-capable image sensors to its popular consumer camera lines, but the IMX500 has already begun shipping to Sony’s industry customers and the IMX501 is set to begin shipping next month, so there’s a likelihood this technology will likely see real-world use very soon.
