This is an on-going unofficial pytorch implementation of the following paper:
Flynn, John, et al. "Quark: Real-time, High-resolution, and General Neural View Synthesis." ACM Transactions on Graphics (TOG) 43.6 (2024): 1-20.
| Status | Milestone | ETA |
|---|---|---|
| 🚀 | Implement all modules that have learnable parameters | 2024-12-31 |
| 🚀 | Implement the train code | TBD |
| 🚀 | Implement the inference code | TBD |
Although I am supposed to implement alone, please let me know if you want to involve in! We may significantly speed the implementation up.
For now this is just for fun and I plan to spend some of my free time, so please also let me know if you want the implementations to be done soon.