Papers club from the AI team in D-ID - this time Neural Networks for Implicit Representations
An implicit neural representations is a parameterization of a signal as a continuous function that maps the domain of the signal to the signal value, for example an float image coordinate to the RGB value.
A really good explnantion on what are "Neural Networks for Implicit Representations" and why are they intresting can be found in awesome-implicit-representations
מועדון קריאת מאמרים שלנו - כל ההרצאות בעיברית
| Title | Paper / Resource | Year | Why is it interesting? | Asignee | Recording | Slides |
|---|---|---|---|---|---|---|
| Neural Fields in Visual Computing and Beyond | Neural Fields in Visual Computing and Beyond | 2022 | read whyWhy and how we can represnt 3d scene using a neural netwrok |
@amitay.nachmani | zoom (d0D9Yv$8) | slides |
| DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation | DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation | 2019 | read whyRepresenting a scene using signed distance functions |
@matan.feldman | zoom (46RYCe*k) | slides |
| Occupancy Networks: Learning 3D Reconstruction in Function Space | Occupancy Networks: Learning 3D Reconstruction in Function Space | 2019 | read whyOccupancy networks implicitly represent the 3D surface as the continuous decision boundary of a deep neural network classifier |
self-work | - | - |
| Implicit Neural Representations with Periodic Activation Functions AKA SIREN | Implicit Neural Representations with Periodic Activation Functions | 2020 | read whySinusoidal representation networks or SIRENs, are ideally suited for representing complex natural signals and their derivatives |
@matan.feldman | zoom (T8Y6@^2N) | slides |
| NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis | NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis | 2020 | read whyThe paper that started all the NERF madness |
@TBH | zoom (kdx=Zd0a) | slides |
| Nerfies: Deformable Neural Radiance Fields | Nerfies: Deformable Neural Radiance Fields | 2021 | read whyPhotorealistically reconstructing deformable scenes using photos/videos captured casually from mobile phone |
@orgoro | zoom (8Z3LeB^Q) | slides |
| Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstruction | Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstruction | 2020 | read whyDynamic neural radiance fields for modeling the appearance and dynamics of a human face |
@ganitk | zoom (CJ+7=!U#) | slides |
| Animatable Neural Radiance Fields from Monocular RGB Videos | Animatable Neural Radiance Fields from Monocular RGB Videos | 2021 | read whyCreating full body avatars using NERF |
@alon.mengi | zoom (Spc7+aYX) | slides |
| I M Avatar: Implicit Morphable Head Avatars from Videos | I M Avatar: Implicit Morphable Head Avatars from Videos | 2022 | read whyCreating an high resolution vavatar only from a cell phone video |
@amitay.nachmani | zoom (*dp9$&%R) | slides |
| ReLU Fields: The Little Non-linearity That Could | ReLU Fields: The Little Non-linearity That Could | 2022 | read whywhat is the smallest change to grid-based representations that allows for retaining the high fidelity result of MLPs while enabling fast reconstruction and rendering times |
@ShiraBaronn | zoom (=8D2Fk1P) | slides |