Depth camera of stereoscopic method for textureless object
Hello.
I`m trying to introduce rgb-d camera to obtain depth values for textureless objects.
In my range of environment, the D405 model is suitable, but the depth camera of the stereoscopic method is theoretically not suitable for textureless objects.
But I tested it on A4 paper with D405 model and got the quite reasonable depth value. I'm happy to get depth value, but I'm confused because it is contrary to my knowledge. How should I accept this result?
-
Hi Leejonghak Although the D405 camera model is not equipped with the dot pattern projector that other RealSense models have to assist depth analysis of surfaces, the 400 Series cameras can alternatively use ambient light in the scene for depth analysis. So if the room lighting is good then that should help to improve the depth results.
-
thank you for your reply. I wonder whether your answer appropriate for texturelss object. In my poor knowledge, stereoscopic method use block matching for gain disparity(and depth). But block matching is not suitable for textureless object because there is no texture.
I wonder if there is something I haven't considered, or overcome obstacles with superior technical skills of realsense.thank you
Please sign in to leave a comment.
Comments
3 comments