Implicit representations are actually very interesting. Forgive me for bringing in Deep Learning stuff into this, but there are some amazing papers very recently on arxiv that are able to do differentiable rendering for implicit functions, which allows for high fidelity 3D reconstruction of objects that have an arbitrary genus (which is not possible with meshes, point clouds, etc. due to some modeling constraints).
These are the papers, if someone is interested:
1) DIST: Rendering Deep Implicit Signed Distance Function
with Differentiable Sphere Tracing (http://b1ueber2y.me/projects/DIST-Renderer/)
2) Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision (https://arxiv.org/abs/1912.07372)
Implicit representations are actually very interesting. Forgive me for bringing in Deep Learning stuff into this, but there are some amazing papers very recently on arxiv that are able to do differentiable rendering for implicit functions, which allows for high fidelity 3D reconstruction of objects that have an arbitrary genus (which is not possible with meshes, point clouds, etc. due to some modeling constraints).
These are the papers, if someone is interested: 1) DIST: Rendering Deep Implicit Signed Distance Function with Differentiable Sphere Tracing (http://b1ueber2y.me/projects/DIST-Renderer/) 2) Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision (https://arxiv.org/abs/1912.07372)