On The Use of Least Expected Features for 3D Keypoint Detection

                                                                            

by Isma Hadji


Introduction


   Object recognition algorithms rely on the detection of a subset of important or discriminative visual stimuli (keypoints) as a first step towards the description of those objects.

Independently of the type of 3D feature used, all 3D detectors rely on a local criterion for keypoint selection. This criterion is usually a point-wise saliency measure that is based on experimentally learnt thresholds. In this research, we question both the threshold based approach, as well as the local characteristic of traditional 3D keypoint detection schemes.

First, we abstract the keypoint selection from experimentally learnt thresholds that depend on low level features for saliency detection. To this end we propose the Least Expected Feature criterion (LEFT) for saliency detection. Second we introduce the concept of finding keypoints considering a global approach, as opposed to more traditional local neighborhood based approaches.

It turns out that adopting the proposed global LEFT criterion allows for the selection of very distinctive keypoints across the entire object, while avoiding sensitive and noisy regions. Examples of the keypoints detected by the Global LEFT detector are shown in Figures 1 and 2, and compared to keypoints detected using the ISS detector. From the figures it is clear that our criteria only selects outstanding points as opposed to the ISS detector that selects points along the edges in the pan example, while it detects keypoints all over the object even in smooth non-important regions on the dragon example.

For details on the implementation of the proposed LEFT detector, the experiments and results we refer the reader to our paper in [1].

 

   

 

    References

  1. I. Hadji and G. N. DeSouza, "On The Use of Least Expected Features for 3D Keypoint Detection ". (Under review for WACV 2015)