Abstract: Partial face recognition (PFR) is challenging as the appearance of the face changes significantly with occlusion. In particular, these occlusions can be due to any item and may appear in any position that seriously hinders the extraction of discriminative features. Existing methods deal with PFR either by training a deep model with existing face databases containing limited occlusion types or by extracting un-occluded features directly from face regions without occlusions. Limited training data (i.e., occlusion type and diversity) can not cover the real-occlusion situations, and thus training-based methods can not learn occlusion robust discriminative features. The performance of occlusion region-based…method is bounded by occlusion detection. Different from limited training data and occlusion region-based methods, we propose to use multi-label attributes for Partial Face Recognition (Attr4PFR). A novel data augmentation is proposed to solve limited training data and generate occlusion attributes. Apart from occlusion attributes, we also include soft biometric attributes and semantic attributes to explore more rich attributes to combat the loss caused by occlusions. To train our Attr4PFR, we propose an implicit attributes loss combined with a softmax loss to enforce Attr4PFR to learn discriminative features. As multi-label attributes are our auxiliary signal in the training phase, we do not need them in the inference. Extensive experiments on public benchmark AR and IJB-C databases show our method is 3% and 2.3% improvement compared to the state-of-the-art.
Show more
Keywords: Partial face recognition, multi-label attributes, discriminative feature learning
Abstract: The goal of this study is to evaluate the feasibility and efficacy of percutaneous transarterial embolization for the treatment of serious renal hemorrhage after renal biopsy. Nine patients with renal hemorrhage had frank pain and gross hematuria as main symptoms after renal biopsy. Intrarenal arterial injuries and perinephric hematoma were confirmed by angiography in all cases. The arterial injuries led to two types of renal hemorrhage, Type I: severe renal injure or intrarenal renal artery rupture (n=5), with contrast medium spilling out of the artery and spreading into renal pelvis or kidney capsule in angiography; Type II, pseudo aneurysm or…potential risk of intrarenal artery injure (n=4), where contrast medium that spilled out of intraartery was retained in the parenchyma as little spots less than 5 mm in diameter in angiography. Transcatheter superselective intrarenal artery embolization was performed with coils or microcoils (Type I intrarenal artery injure) and polyvinyl alcohol particles (Type II injure). The intrarenal arterial injuries were occluded successfully in all patients. Light or mild back or abdominal pain in the side of the embolized kidney was found in three patients following embolization procedures and disappeared 3 days later. Serum creatinine and perinephric hematoma were stable, and gross hematuresis stopped immediately (n=4) or 3–5 days (n=3) after embolization. In conclusions, transcatheter superselective intrarenal artery embolization as a minimally invasive therapy is safe and effective for treatment of serious renal hemorrhage following percutaneous renal biopsy.
Show more