Unsupervised Salience Learning for Person Re-identification CVPR2013 Poster
Outline Introduction Method Experiments Conclusions
Introduction Human eyes can recognize person identities based on some small salient regions.
Introduction Person re-identification handles pedestrian matching and ranking across non-overlapping camera views. viewpoint change and pose variation cause uncontrolled misalignment between images.
Introduction Motivations : 1. We can recognize persons across camera views from their local distinctive regions 2. Human salience 3. Distinct patches are considered as salient only when they are matched and distinct in both camera views
Method Dense Correspondence Unsupervised Salience Learning Matching for Re-identification
Dense Correspondence Features: Adjacency constrained search: dense color histogram + dense SIFT Adjacency constrained search: simple patch matching
Adjacency constrained search Search set :
Adjacency constrained search Adjacency Searching:
Unsupervised Salience Learning two methods for learning human salience: K-Nearest Neighbor Salience (KNN) One-Class SVM Salience (OCSVM)
Unsupervised Salience Learning Definition: Salient regions are discriminative in making a person standing out from their companions, and reliable in finding the same person across camera views. Assumption: fewer than half of the persons in a reference set share similar appearance if a region is salient. Hence, we set k = Nr/2. Nr is the number of images in reference set.
Unsupervised Salience Learning
K-Nearest Neighbor Salience
K-Nearest Neighbor Salience
Unsupervised Salience Learning
One-Class SVM Salience
Matching for Re-identification Bi-directional Weighted Matching Complementary Comination
Matching for Re-identification
Experiments Dataset : VIPeR Dataset ETHZ Dataset
Experiments
Experiments
Experiments
Experiments
Experiments
Conclusions 1. An unsupervised framework to extract distinctive features for person re-identification. 2. Patch matching is utilized with adjacency constraint for handling the misalignment problem caused by viewpoint change and pose variation. 3. Human salience is learned in an unsupervised way.