Abstract: Many unsupervised domain adaptation (UDA) methods have been proposed to bridge the domain gap by utilizing domain invariant information. Most approaches have chosen depth as such information and achieved remarkable successes. Despite their effectiveness, using depth as domain invariant information in UDA tasks may lead to multiple issues, such as excessively high extraction costs and difficulties in achieving a reliable prediction quality. As a result, we introduce Edge Learning based Domain Adaptation (ELDA), a framework which incorporates edge information into its training process to serve as a type of domain invariant information. Our experiments quantitatively and qualitatively demonstrate that the incorporation of edge information is indeed beneficial and effective, and enables ELDA to outperform the contemporary state-of-the-art methods on two commonly adopted benchmarks for semantic segmentation based UDA tasks.

Abstract: A shock absorbing protection structure for a handheld tool contains: a power tool, a shock absorption sleeve, and at least one stopper. The power tool includes a drive block which has a switch configured to turn on/off the drive block, the driving block also has a coupling tube which has a tool head fixed on the coupling tube away from the drive block, and the power tool includes a joining portion formed on the power tool. The shock absorption sleeve is fitted on the joining portion of the power tool, and between the shock absorption sleeve and the joining portion is defined an accommodation space. The at least one stopper is flexible and is defined between the shock absorption sleeve and the joining portion of the power tool, and the at least one stopper is located at the accommodation space so as to absorb vibration.