A Largest Matching Area Approach to Image Denoising

Given the success of patch-based approaches to image denoising,this paper addresses the ill-posed problem of patch size selection.Large patch sizes improve noise robustness in the presence of good matches, but can also lead to artefacts in textured regions due to the rare patch effect; smaller patch sizes reconstruct details more accurately but risk over-fitting to the noise in uniform regions. We propose to jointly optimize each matching patch’s identity and size for gray scale image denoising, and present several implementations.The new approach effectively selects the largest matching areas, subject to the constraints of the available data and noise level, to improve noise robustness. Experiments on standard test images demonstrate our approach’s ability to improve on fixed-size reconstruction, particularly at high noise levels, on smoother image regions.

Person Re-Identification using Deep Convnets with Multi-task Learning

Person re-identification involves recognizing a person across non-overlapping camera views, with different pose, illumination, and camera characteristics. We propose to tackle this problem by training a deep convolutional network to represent a person’s appearance as a low-dimensional feature vector that is invariant to common appearance variations encountered in the re-identification problem. Specifically, a Siamese-network architecture is used to train a feature extraction network using pairs of similar and dissimilar images. We show that use of a novel multi-task learning objective is crucial for regularizing the network parameters in order to prevent over-fitting due to the small size the training dataset. We complement the verification task, which is at the heart of re-identification, by training the network to jointly perform verification, identification, and to recognise attributes related to the clothing and pose of the person in each image. Additionally, we show that our proposed approach performs well even in the challenging cross-dataset scenario, which may better reflect real-world expected performance. 

Size-fitting of Intravaginal Rings for Macaques and in vitro Release Kinetics of Zinc Finger Inhibitors

Background: Small molecule inhibitors of the zinc finger domain (ZFI) in the nucleocapsid protein (NCp7) of HIV-1 are potent inhibitors of HIV and SIV
replication and may have utility as topical products to prevent infection. Furthermore, intravaginal rings (IVRs) were developed as coitally-independent,
sustained release devices which could be used for administration of HIV microbicides. The aims of these studies were to demonstrate that IVRs sized for
macaques are practical and compatible with the current generation of thioester-based NCp7 inhibitors.

Methods: Non-medicated silicone elastomer vaginal rings of various sizes thought to be applicable for macaques were prepared and tested for vaginal fit in Pigtailed and Chinese Rhesus macaques. Macaques were monitored for 8 weeks for mucosal disruption by colposcopy and proinflammatory cytokine markers in cervical vaginal lavages (CVL) using Luminex bead-based technology. Three different ZFIs (compounds 52, 89 and 122, each derived from an N-substituted S-acyl-2-mercaptobenzamide thioester scaffold) were loaded at 50 mg into an optimal matrix-type ring design. In vitro continuous release studies were then conducted over 28 days and analyzed by HPLC. Rate of release was determined by linear regression analysis.

Results: Qualitative evaluation at the time of ring insertion suggested that the 25 mm ring provided optimal fit in both macaque species. All rings remained in
place during the study period (2 to 4 weeks), and the animals did not attempt to remove the rings. No tissue irritation was observed, and no signs of physical
discomfort were noted. Also, no significant induction of cervicovaginal proinflammatory markers was observed during the 8-week period during and following ring insertion. One Pigtailed macaque showed elevated IL-8 levels in the CVL during the period when the ring was in place; however, these levels were comparable to those observed in two control macaques. In vitro release of the ZFIs peaked at day 1 and then continually declined to near steady-state rates between 20-30 mcg/day. The percent release after 14 days was 2.9, 2.0 and 0.9 for ZFI 89, 52 and 122, respectively.

Conclusions: IVRs of 25mm diameter, determined to be the optimal size for macaques, were well tolerated and did not induce inflammation. Release of all ZFI compounds followed t 0.5 kinetics. These findings suggest that efficacy testing in primate models is warranted to fully evaluate the potential to prevent
transmission.