国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁(yè)> 外文會(huì)議>IEEE/CVF Conference on Computer Vision and Pattern Recognition >Hierarchical Novelty Detection for Visual Object Recognition
【24h】

Hierarchical Novelty Detection for Visual Object Recognition

機(jī)譯：視覺對(duì)象識(shí)別的層次新穎性檢測(cè)

獲取原文
  • 摘要
  • 著錄項(xiàng)
  • 相似文獻(xiàn)
  • 相關(guān)主題

摘要

Deep neural networks have achieved impressive success in large-scale visual object recognition tasks with a predefined set of classes. However, recognizing objects of novel classes unseen during training still remains challenging. The problem of detecting such novel classes has been addressed in the literature, but most prior works have focused on providing simple binary or regressive decisions, e.g., the output would be 'known,' 'novel,' or corresponding confidence intervals. In this paper, we study more informative novelty detection schemes based on a hierarchical classification framework. For an object of a novel class, we aim for finding its closest super class in the hierarchical taxonomy of known classes. To this end, we propose two different approaches termed top-down and flatten methods, and their combination as well. The essential ingredients of our methods are confidence-calibrated classifiers, data relabeling, and the leave-one-out strategy for modeling novel classes under the hierarchical taxonomy. Furthermore, our method can generate a hierarchical embedding that leads to improved generalized zero-shot learning performance in combination with other commonly-used semantic embeddings.
機(jī)譯：深度神經(jīng)網(wǎng)絡(luò)在具有預(yù)定義類集的大規(guī)模視覺對(duì)象識(shí)別任務(wù)中取得了令人矚目的成功。然而，識(shí)別在訓(xùn)練期間看不見的新穎類的物體仍然是挑戰(zhàn)。在文獻(xiàn)中已經(jīng)解決了檢測(cè)這種新穎類的問題，但是大多數(shù)先前的工作集中在提供簡(jiǎn)單的二進(jìn)制或回歸決策上，例如，輸出將是“已知的”，“新穎的”或相應(yīng)的置信區(qū)間。在本文中，我們研究基于分層分類框架的更多信息新穎性檢測(cè)方案。對(duì)于小說(shuō)類的對(duì)象，我們旨在在已知類的層次分類法中找到其最接近的超類。為此，我們提出了兩種不同的方法，稱為自頂向下和展平方法，以及它們的組合。我們方法的基本要素是置信度校準(zhǔn)的分類器，數(shù)據(jù)重新標(biāo)記以及在分層分類法下為新型類建模的留一法。此外，我們的方法可以生成分層嵌入，從而與其他常用語(yǔ)義嵌入結(jié)合使用，從而提高廣義零擊學(xué)習(xí)性能。

著錄項(xiàng)

相似文獻(xiàn)

  • 外文文獻(xiàn)
  • 中文文獻(xiàn)
  • 專利
獲取原文

客服郵箱：kefu@zhangqiaokeyan.com

京公網(wǎng)安備：11010802029741號(hào) ICP備案號(hào)：京ICP備15016152號(hào)-6 六維聯(lián)合信息科技 (北京) 有限公司?版權(quán)所有

  • 客服微信

  • 服務(wù)號(hào)