search

microsoft / AirSim-NeurIPS2019-Drone-Racing

Drone Racing @ NeurIPS 2019, built on Microsoft AirSim
https://microsoft.github.io/AirSim-NeurIPS2019-Drone-Racing/
MIT License
356 stars 89 forks source link

Infrared data #138

Open collinabidi opened 4 years ago

collinabidi commented 4 years ago

I'm using the baseline_racer_image_benchmarker.py to test the speed of capturing images from cameras with different sensor types. However, when I try to get Segmentation/Infrared/Surface Normal data, I get color images.

I have correct camera/capture settings in the settings.json file, so I'm not worried about that.

Is there something I'm missing when I use the following request to get both RGB and Infrared data?

request = [airsim.ImageRequest("fpv_cam", airsim.ImageType.Scene, False, False),airsim.ImageRequest("fpv_cam", 7, False, False)]

I've tried changing the float and compress parameters to see what happens but the second requested image still results in color images.

collinabidi commented 4 years ago

Issue with getting depth data was solved, but I'm confused about the differences between all of the image types. The Python API documentation has a lot of placeholders; could someone explain how the different image types are formatted?