English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, we study the task of detecting semantic parts of an object,e.g., a wheel of a car, under partial occlusion. We propose that all modelsshould be trained without seeing occlusions while being able to transfer thelearned knowledge to deal with occlusions. This setting alleviates thedifficulty in collecting an exponentially large dataset to cover occlusionpatterns and is more essential. In this scenario, the proposal-based deepnetworks, like RCNN-series, often produce unsatisfactory results, because boththe proposal extraction and classification stages may be confused by theirrelevant occluders. To address this, [25] proposed a voting mechanism thatcombines multiple local visual cues to detect semantic parts. The semanticparts can still be detected even though some visual cues are missing due toocclusions. However, this method is manually-designed, thus is hard to beoptimized in an end-to-end manner.  In this paper, we present DeepVoting, which incorporates the robustness shownby [25] into a deep network, so that the whole pipeline can be jointlyoptimized. Specifically, it adds two layers after the intermediate features ofa deep network, e.g., the pool-4 layer of VGGNet. The first layer extracts theevidence of local visual cues, and the second layer performs a voting mechanismby utilizing the spatial relationship between visual cues and semantic parts.We also propose an improved version DeepVoting+ by learning visual cues fromcontext outside objects. In experiments, DeepVoting achieves significantlybetter performance than several baseline methods, including Faster-RCNN, forsemantic part detection under occlusion. In addition, DeepVoting enjoysexplainability as the detection results can be diagnosed via looking up thevoting cues.

DeepVoting: A Robust and Explainable Deep Network for Semantic Part Detection under Partial Occlusion