Visual content description is a key issue for the task of machine-based visual object categorization (VOC). A good visual descriptor should be both discriminative enough and computationally efficient while possessing some properties of robustness to viewpoint changes and lighting condition variations. The recent literature has featured local image descriptors, e.g. SIFT (Scale Invariant Feature Transform), as the main trend in VOC. However, it is well known that SIFT is computationally expensive and hardly scales when the number of objects/concepts and learning data increase significantly as the case in TRECVID. In this paper, we investigate the DAISY, which is a new fast local descriptor first introduced for wide baseline matching problem, in the context of visual object recognition. We carefully evaluate and compare the DAISY descriptor with SIFT both in terms of performance and time complexity on two standard image benchmarks – Caltech 101 and PASCAL VOC 2007. The experimental results show that DAISY, while using shorter descriptor length and operating 3 times faster, displays a better recognition accuracy than SIFT. When displaying a similar recognition accuracy to SIFT, DAISY can operate 12 times faster.