作者: Paul Schmieder , Beryl Plimmer , Rachel Blagojevic
关键词:
摘要: We present our toolkit to automatically evaluate recognition algorithms. There are few published comparative evaluations of sketch algorithms and those that exist do not provide benchmarking or direct comparisons because standardised data an evaluation platform is available. By unifying collection, labelling in one tool, fair, flexible comprehensive possible. Currently we have 6 existing recognizers integrated into this tool. With initial these observed the context from which training taken has effect on success rates. These results suggest such as a powerful adjunct for research.