Learned Initializations for Neural Representations

Learned Initializations for Optimizing Coordinate-Based Neural Representations

Learned Initializations for Neural Representations Learned Initializations for Optimizing Coordinate-Based Neural Representations CVPR 2021 Matthew Tancik * 1* Ben Mildenhall * 1* Terrance Wang 1 Divi Schmidt 1 UC Berkeley UC Berkeley UC Berkeley UC Berkeley Pratul P. Srinivasan 2 Jonathan T. Barron 2 Ren Ng 1 Matthew Tancik * 1* Ben Mildenhall * 1* Terrance Wang 1 Divi Schmidt 1 Pratul P. Srinivasan 2 Jonathan T. Barron 2 Ren Ng 1 1 UC Berkeley 2 Google Research * * Equal Contribution Google Research Google Research UC Berkeley * * Denotes Equal Contribution Paper Code Data Overview Coordinat

related reading

• NeRF: Neural Radiance Fieldsmatthewtancik.com
• Fourier Feature Networksbmild.github.io
• NL.pdfabehrouz.github.io
• Feature Visualizationdistill.pub
• MetaSDF: Meta-learning Signed Distance Functionsarxiv.org
• [2205.13147] Matryoshka Representation Learningarxiv.org
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• nn-notes.pdfboris-hanin.github.io
• Understanding deep learning requires rethinking generalizationarxiv.org
• Naturally Occurring Equivariance in Neural Networksdistill.pub
• 2006.10742arxiv.org
• Multimodal Neurons in Artificial Neural Networksdistill.pub