In principle, an extra sensor that measures red and near-infrared light could give the camera a rough clue about the type of illumination. Different light sources have different spectral signatures, so comparing red vs. near-IR intensity might help distinguish broad categories such as daylight, incandescent, or some artificial lighting, which could improve auto white balance.

That said, white balance is not determined by a single universal measurement. It’s partly an estimation problem, and cameras can often do it in software by analyzing the image data itself. That’s why RAW converters can also suggest white balance after the fact.

DSLRs and other dedicated cameras generally do not rely on an IR heart-rate style sensor for white balance. They more commonly use the main image sensor, metering sensor, scene analysis, presets, or user input. Many cameras also let you set color temperature manually and apply green-magenta correction.

So the idea is plausible as an assistive input, but it’s probably just one clue among others rather than a complete scientific solution to white balance.