Koen Claessen in the Lava project has some interesting publications.

Analysis of Programs for Parallel Processing, in 1966, is a pioneering work on dependency analysis of program.

Some recent papers from Computer Vision and Image Understanding:

• Design and implementation of embedded computer vision systems based on particle filters
• Leveraging Cost Matrix Structure for Hardware Implementation of Stereo Disparity Computation Using Dynamic Programming

Off topic:
Thoughts on Flash, by Steve Jobs, April 29th, 2010.
HiPEAC collaboration Grants is giving away some money.

IEEE Workshop on Embedded Computer Vision 2010 releases its program. Every work is interesting.

A Dynamically Reconfigurable Architecture Combining Pixel-Level SIMD and Operation-Pipeline Modes for High Frame Rate Visual Processing, in FPT’2007, from the Hashimoto-Kagami Lab of Tohoku University.

Some recent papers in Computer Vision and Image Understanding:

• A Parallel Histogram-based Particle Filter for Object Tracking on SIMD-based Smart Cameras, from Purdue. They still use WiCa.
• Wireless smart camera network for real-time human 3D pose reconstruction
• Light-weight salient foreground detection for embedded smart cameras

Xilinx puts the ARM hard core into the road map.

A Graphics Parallel Memory Organization Exploiting Request Correlations, in TC June 2010, shows an example of modeling memory partition and banking. The title is on graphics, but the method also applies to vision. This work shows that can be further improved on architectural template. The exploitation of request correlations can also be further improved.

The proceeding of CGO 2010 is available. Exciting moment. The conference is on April 24 – 28, 2010, but the proceeding is released before April 28. The ACM digital library is damn quick.

ODES: Workshop on Optimizations for DSP and Embedded Systems, co-located with CGO 2010, is also interesting. The program and proceeding of ODES 2010 is available from the workshop website.

Two papers in CGO 2010 are on the topic of tiling for autotuning:

• Parameterized tiling revisited, by Muthu Manikandan Baskaran from OSU, who has done good work on autuning for accelerator architecture.
• Automatic creation of tile size selection models, from IBM. The Tile-Size Selection (TSS) model is a nice example for modeling the system.

Some links for Prolog:

• Prolog wikibook.
• Prolog Tutorial by J.R.Fisher.
• SWI-Prolog manual
• win-prolog documentation.

special issue MICPRO journal on exploitation of hardware accelerators, deadline on September 10, 2010.

EuroSys 2010 has some papers on schedulers:

• Bias scheduling in heterogeneous multi-core architectures
• A comprehensive scheduler for asymmetric multicore systems
• Resource-conscious scheduling for energy efficiency on multicore processors

The May 2010 issue of TVLSI has some interesting papers:

• Design Space Exploration of Hard-Decision Viterbi Decoding: Algorithm and VLSI Implementation, their design space exploration mainly focus on dedicated building blocks, implemented in ASIC instead of FPGA.
• Low-Power Multimedia System Design by Aggressive Voltage Scaling, a lot of estimation technique is used.
• A Discussion on SRAM Circuit Design Trend in Deeper Nanometer-Scale Technologies, survey-like paper.

Some interesting papers in June 2010 issue of TPAMI:

• Correction of Spatially Varying Image and Video Motion Blur Using a Hybrid Camera, one high resolution slow camera and one low resolution fast camera.
• Multibody Structure-from-Motion in Practice
• The Improbability of Harris Interest Points

Some papers on the Auto-Pipe tool, which is for developing streaming application in hybrid systems, namely architectural diverse systems, including CPU, FPGAs, and GPUs. It is related work. However, it requires manual implementation of kernels on target architectures.

• Application-guided Tool Development for Architecturally Diverse Computation, in Symposium on Applied Computing (SAC) 2010. It summaries the lessons learn from the Auto-Pipe tool. One lesson learned is the inadequacy of synchronous data flow graph to specify run time dynamic word load, e.g. input dependent workload. Another lesson learned is the signalling protocol between components.
• Auto-Pipe: Streaming Applications on Architecturally Diverse Systems, in IEEE Computer March 2010. An introduction to Auto-Pipe.
• Visions for application development on hybrid computing systems, in Journal of Parallel Computing 2008, calling for a proper description language for concurrency. Instead of describing concurrency, I prefer to describe dependency. Dependency decides concurrency.
• More papers on auto-pipe from Roger Chamberlain’s publication list.

Massively parallel computing on an organic molecular layer, in Nature Physics 2010 April. It looks cool, although I have not idea what that paper is talking about. It is unlikely to be commercially viable before I die.

Some publications on high level languages for image processing from QUB:

• The “TULIP” — An abstract machine approach to environments for image interpretation on transputers, in 1994.
• The “IAL” — IAL: a parallel image processing programming language, in 1990.
• The “PIPL” — A portable abstract machine model for image processing: an implementation technique for software tools, in 1993.
• The “I-BOL” — A high level language for parallel image processing, in 1994.

For an application with tight delay and throughput requirements, it would be nice to have a chart showing the application’s latency and throughput on different architectures. For visual servoing application, such chart would be interesting.

Off topic:
A Ph.D. is not enough: a guide to survival in science, showing the dark but real side of the academic world.

stanford-cs193g-sp2010: “Programming Massively Parallel Processors with CUDA”, a new GPU course in Stanford.

Off topic:
Yet another homepage of mine.