We present a generic detection/localization algorithm capable of searching for a visual object of interest without training. The proposed method operates using a single example of an object of interest to find similar matches; does not require prior knowledge (learning) about objects being sought; and does not require any pre-processing step or segmentation of a target image. Our method is based on the computation of local regression kernels as descriptors from a query, which measure the likeness of a pixel to its surroundings. Salient features are extracted from said descriptors and compared against analogous features from the target image. This comparison is done using a matrix generalization of the cosine similarity measure. We illustrate optimality properties of the algorithm using a naive-Bayes framework. The algorithm yields a scalar resemblance map, indicating the likelihood of similarity between the query and all patches in the target image. By employing nonparametric significance tests and non-maxima suppression, we detect the presence and location of objects similar to the given query. The approach is extended to account for large variations in scale and rotation. High performance is demonstrated on several challenging datasets, indicating successful detection of objects in diverse contexts and under different imaging conditions.

System Overview

Results on the UIUC car test set

Results on the MIT-CMU face test set

Results on Shechtman general objects test set

OpenCV Implementation link

Note: This code requires ROS to be installed first.(ROS is an open-source, meta-operating system for your robot.) : installation guide

Matlab Package

Disclaimer: This is experimental software. It is provided for non-commercial research purposes only. Use at your own risk. No warranty is implied by this distribution. Copyright © 2010 by University of California.