fabiotosi92/NeRF-Supervised-Deep-Stereo

how generate center view on my stereo pairs scene ?

565ee opened this issue · 6 comments

565ee commented

thank your NeRF-Supervised-Deep-Stereo !! i like it very much !!

but now, i have a question :
i have left.jpg right.jpg disp0.pfm calib.txt
and use colmap generate : poses\colmap_sparse colmap_text colmap.db

i use : python test.py ... , generate .../outdir/0.jpg 0.npy
i use : python demo.py ... , generate .../outdir/disparity_map.png

now , how generate center view on my stereo pairs scene by baseline ?
what code? what python packages?
image

fabiotosi92 commented

Hi, are you trying to generate stereo pairs on your specific scene or are you using our dataset? In the latter case, you already have the center view for each rendered triplet. If you need more information, please refer to the documentation, which you can find at this link.

565ee commented

how generate center view on my stereo pairs scene by baseline ?

fabiotosi92 commented

Once you have chosen a specific viewpoint, you can render the center image. From there, you can compute the pose by starting with an arbitrary baseline. This allows you to synthesize the left and right views. By doing so, you have your desired triplet.

565ee commented

what code? what python packages?

fabiotosi92 commented

@565ee Pay attention to the fact that in our work, we do not generate a central view from a given stereo pair. Our approach involves generating stereo pairs from a NeRF model trained on multiple images captured by a single camera from different viewpoints of the scene.