| 1. |
- Okada, Ryuzo, et al.
(författare)
-
Chairs message
- 2019
-
Ingår i: Proceedings of the 16th International Conference on Machine Vision Applications, MVA 2019. - : Institute of Electrical and Electronics Engineers Inc..
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)
|
|
| 2. |
|
|
| 3. |
- Shirai, Takaaki, et al.
(författare)
-
Longitudinal evaluation of cartilage after osteochondral autogenous transfer with delayed gadolinium-enhanced MRI of the cartilage (dGEMRIC)
- 2012
-
Ingår i: Journal of Orthopaedic Research. - : Wiley. - 1554-527X .- 0736-0266. ; 30:2, s. 221-225
-
Tidskriftsartikel (refereegranskat)abstract
- The aim was to use repeat delayed gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC) to estimate glycosaminoglycan (GAG) content in reparative cartilage after osteochondral autogenous transfer (OAT). The study group comprised 7 knees of 7 patients that were examined three times by dGEMRIC, at 3, 6, and 12 months using a 1.5 Tesla MRI system in both OAT operated and nonoperated condyles at 90 min after the injection. The gadolinium diethylene triamine pentaacetic acid (Gd-DTPA)2- containing contrast medium (0.2 mmols/kg) was injected intravenously. The mean T1 values of the plug cartilage at 3, 6, and 12 months after OAT was 230 +/- 40, 213 +/- 31, and 230 +/- 23 ms (mean +/- SD), respectively. There were differences between the plug and control cartilage at 3 (p<0.01) and 12 (p<0.05) months after OAT, but not at 6 months (p=0.089). No T1 changes were detected between the plug cartilage at the different time points after OAT. The fact that the GAG content of the OAT plugs were maintained for 12-month study period suggest that no major deterioration of load-bearing properties occurs in the cartilage after the OAT. (C) 2011 Orthopaedic Research Society Published by Wiley Periodicals, Inc. J Orthop Res 30:221225, 2012
|
|
| 4. |
- Yang, Zhong, et al.
(författare)
-
Target aware network adaptation for efficient representation learning
- 2018
-
Ingår i: ECCV 2018: Computer Vision – ECCV 2018 Workshops. - Munich : Springer. - 9783030110178 ; , s. 450-467
-
Konferensbidrag (refereegranskat)abstract
- This paper presents an automatic network adaptation method that finds a ConvNet structure well-suited to a given target task, e.g. image classification, for efficiency as well as accuracy in transfer learning. We call the concept target-aware transfer learning. Given only small-scale labeled data, and starting from an ImageNet pre-trained network, we exploit a scheme of removing its potential redundancy for the target task through iterative operations of filter-wise pruning and network optimization. The basic motivation is that compact networks are on one hand more efficient and should also be more tolerant, being less complex, against the risk of overfitting which would hinder the generalization of learned representations in the context of transfer learning. Further, unlike existing methods involving network simplification, we also let the scheme identify redundant portions across the entire network, which automatically results in a network structure adapted to the task at hand. We achieve this with a few novel ideas: (i) cumulative sum of activation statistics for each layer, and (ii) a priority evaluation of pruning across multiple layers. Experimental results by the method on five datasets (Flower102, CUB200-2011, Dog120, MIT67, and Stanford40) show favorable accuracies over the related state-of-the-art techniques while enhancing the computational and storage efficiency of the transferred model.
|
|
