Hey, I wonder what is your thoughts on the image quality of the saved images. It's 360x160 per image and looking like
While it's most likely ok for end-to-end training/segmentation, traffic lights are not very distinguishable. How'd we work with this? Should we increase the quality?
Hey, I wonder what is your thoughts on the image quality of the saved images. It's 360x160 per image and looking like
While it's most likely ok for end-to-end training/segmentation, traffic lights are not very distinguishable. How'd we work with this? Should we increase the quality?