ABSTRACT
Describing the color and textural information of a person image is one of the most crucial aspects of person re-identification (re-id). Although a covariance descriptor has been successfully applied to person re-id, it loses the local structure of a region and mean information of pixel features, both of which tend to be the major discriminative information for person re-id. In this paper, we present novel meta-descriptors based on a hierarchical Gaussian distribution of pixel features, in which both mean and covariance information are included in patch and region level descriptions. More specifically, the region is modeled as a set of multiple Gaussian distributions, each of which represents the appearance of a local patch. The characteristics of the set of Gaussian distributions are again described by another Gaussian distribution. Because the space of Gaussian distribution is not a linear space, we embed the parameters of the distribution into a point of Symmetric Positive Definite (SPD) matrix manifold in both steps. We show, for the first time, that normalizing the scale of the SPD matrix enhances the hierarchical feature representation on this manifold. Additionally, we develop feature norm normalization methods with the ability to alleviate the biased trends that exist on the SPD matrix descriptors. The experimental results conducted on five public datasets indicate the effectiveness of the proposed descriptors and the two types of normalizations.
ABSTRACT
PURPOSE: Traditionally and since Neer, the humeral side of shoulder arthroplasty consisted of a stemmed component but the real need for stem fixation in total shoulder arthroplasty (TSA) has barely been investigated. The current study evaluated the clinical and radiological outcomes with a stemless TSA. METHODS: Forty-seven patients, 20 female and 27 male patients with an average age of 63, were selected in four orthopaedic centres during a four year period, and implanted with a humeral head prosthesis with a three-fin design and titanium coating. Aetiologies were: primary osteoarthritis (29), fracture sequelae (12) and avascular osteonecrosis (6). Minimum follow-up was two years (range 24-51 months). The patients were evaluated with the Constant score (CS) and radiological exams. RESULTS: Two patients had revision of the implants, one for persistent pain and one for secondary massive rotator cuff tear. At the final follow-up, the mean CS was 69, with an average gain of 36. All parameters improved with a foremost in pain relief. Mean satisfaction rate was 87%. Average anterior active elevation was 131° with a gain of 48. Radiologic evaluations showed stable implants in all cases. However, 17 cases demonstrated radiolucent areas, particularly superior and lateral to the implant, which neither decreased nor increased with time. No revisions were related to humeral component loosening. CONCLUSIONS: Stemless TSA provides the same results as compared to TSA with a humeral stem. We are still unsure as to the nature of the lucent zones and we are continuing our investigation to better understand this radiological phenomenon.
