Brennan Shacklett

I am a Ph.D. student at Stanford advised by Kayvon Fatahalian and working on problems in the intersection between graphics and systems. Specifically, I am interested in leveraging the capabilities of modern GPUs (such as their dedicated deep learning hardware) to bring complex rendering techniques to real-time contexts. More broadly, I am also interested in developing debugging and analysis infrastructure to make visual computing systems easier to build.

Current Research

Continuous Learning for Real-Time Denoising

Denoising path traced effects with a convolutional neural network that continually specializes to the current on-screen content.

Past Research

Interactive Circuit Debugging

Undergraduate Honors Thesis: Created a cycle accurate circuit simulator and accompanying debugger, which leverage JIT compilation to achieve high simulation performance while still allowing unrestricted access to intermediate values and state.

Interactive Circuit Debugging and Simulation with Just In Time Compilation

Brennan Shacklett, Pat Hanrahan

Source code

ExCamera

Built a state tracking system to isolate serial dependencies in video streams. Also explored techniques to rapidly change quality levels while streaming.

Encoding, Fast and Slow: Low-Latency Video Processing Using Thousands of Tiny Threads

S Fouladi, R Wahby, B Shacklett, K Balasubramaniam, W Zeng, R Bhalerao, A Sivaraman, G Porter, and K Winstein

Source code