Adam T. Lake

Larrabee: A Many-Core x86 Architecture for Visual Computing

The Larrabee many-core visual computing architecture uses multiple in-order x86 cores augmented by wide vector processor units, together with some fixed-function logic. This increases the architectures programmability as compared to standard GPUs. The article describes the Larrabee architecture, a software renderer optimized for it, and other highly parallel applications. The article analyzes performance through scalability studies based on real-world workloads.

AdaPT: Real-time adaptive pedestrian tracking for crowded scenes

We present a novel realtime algorithm to compute the trajectory of each pedestrian in a crowded scene. Our formulation is based on an adaptive scheme that uses a combination of deterministic and probabilistic trackers to achieve high accuracy and efficiency simultaneously. Furthermore, we integrate it with a multi-agent motion model and local interaction scheme to accurately compute the trajectory of each pedestrian. We highlight the performance and benefits of our algorithm on well-known datasets with tens of pedestrians.

Rigid body collision detection on the GPU

We propose a technique for collision detection of rigid bodies using cube-maps. This technique runs entirely on todays GPUs. Conventional rigid body collision detection algorithms run on the CPU. This technique is highly parallel in nature and highly scalable. At the core of the algorithm is the idea of creating a cube-map for each unique object. This cube-map stores the distances to the surface of the object from its centroid. The algorithm has two steps - a preprocessing step, and a run-time step. Preprocessing is done only once per unique object and the run-time step is evaluated for every instance of the object every frame.

FasTC: accelerated fixed-rate texture encoding

We present a new algorithm for encoding low dynamic range images into fixed-rate texture compression formats. Our approach provides orders of magnitude improvements in speed over existing publicly-available compressors, while generating high quality results. The algorithm is applicable to any fixed-rate texture encoding scheme based on Block Truncation Coding and we use it to compress images into the OpenGL BPTC format. The underlying technique uses an axis-aligned bounding box to estimate the proper partitioning of a texel block and performs a generalized cluster fit to compute the endpoint approximation. This approximation can be further refined using simulated annealing. The algorithm is inherently parallel and scales with the number of processor cores. We highlight its performance on low-frequency game textures and the high frequency Kodak Test Image Suite.

OpenCL™ FFT Optimizations for Intel® Processor Graphics

In this paper, we explore a number of OpenCL™ optimization strategies and show the pros and cons relative to clFFT, the leading OpenCL Fast Fourier Transform (FFT) library. We implemented a 1D, multi-kernel, mixed-radix Cooley-Tukey power of two algorithms that improves upon clFFT for many cases under consideration. The computation is broken down into a set of auto-generated smaller-base FFTs that fit in the execution unit registers, avoiding the use of local memory and barriers; our implementation achieves high-thread occupancy and high memory bandwidth for a wide range of global and local sizes.