Novel View Synthesis: Comparisons to Baselines

We compare Total-Recon (ours) to HyperNeRF, D²NeRF, and their depth-supervised equivalents on novel-view synthesis for RGBD sequences captured with a stereo validation rig we have built. While the baseline methods are only able to reconstruct the background at best, our method is able to reconstruct both the background and the moving foreground(s), demonstrating holistic scene reconstruction.

	Novel View (GT)	Ours (w/ depth)	D2NeRF (w/ depth)	D2NeRF (w/o depth)	HyperNeRF (w/ depth)	HyperNeRF (w/o depth)
Human 1 & Dog 1
Human 2 & Cat 1
Dog 1 (v1)
Dog 1 (v2)
Cat 1 (v1)
Cat 1 (v2)
Cat 2 (v1)
Cat 2 (v2)
Cat 3
Human 1
Human 2