Mobile product recognition with efficient Bag-of-Phrase visual search

Content Based Image Retrieval and Its Application to Product Recognition

“…Product Image Retrieval (PIR) is an emerging field of content-based image retrieval (CBIR). Several algorithms and systems have been introduced in literature for this new research topic [4][5][6][7][8][9][10]. From these studies it can be seen that the product recognition problem falls into two categories: a) systems where users get online information regarding a product by snapping a picture of a product with a camera-device (smartphone or tablet) and b) systems where a vertical search is performed based on domain-specific search such as "Philips TVs" domain.…”

Feature weighting with augmented visual phrase in visual product recognition

“…To handle spurious matches and obtain good match, we apply GV to top 50 ranked images. The performance comparison of the proposed AVP method with respect to the developed BoP method in [10] and the implemented SVT method in [8] is shown in Table. 4.1.…”

“…Further, the proposed method outperforms the efficient BoP by 7.8%. Even though the efficient BoP method in [10] uses spatial information, it cannot handle query images with high photometric and strong geometric distortions. Overall, the proposed method can address these issues and achieves a good performance.…”

“…(1) Spatial information to make features more discriminative; (2) Augmented features to handle potential distortion, including different illumination and perspective changes. Experimental results on a commercial product dataset shows the proposed AVP method achieves a recognition rate of 93.1%, which outperforms the current BoP method [10] by 9%.…”

“…Particularly in BoP approaches, visual words co-occuring within certain spatial region are selected as discriminative visual phrases (DVP) in [20], [22] and [10]. However, these methods only consider the original database images for DVP candidate generation and selection.…”