TUdatalib : Actor-critic Instance Segmentation

Zur Kurzanzeige

dc.contributor.author	Araslanov, Nikita
dc.contributor.author	Rothkopf, Constantin
dc.contributor.author	Roth, Stefan
dc.date.accessioned	2021-12-22T11:09:38Z
dc.date.available	2021-12-22T11:09:38Z
dc.date.issued	2019-06
dc.identifier.uri	https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/3368
dc.description	Most approaches to visual scene analysis have emphasised parallel processing of the image elements. However, one area in which the sequential nature of vision is apparent, is that of segmenting multiple, potentially similar and partially occluded objects in a scene. In this work, we revisit the recurrent formulation of this challenging problem in the context of reinforcement learning. Motivated by the limitations of the global max-matching assignment of the ground-truth segments to the recurrent states, we develop an actor-critic approach in which the actor recurrently predicts one instance mask at a time and utilises the gradient from a concurrently trained critic network. We formulate the state, action, and the reward such as to let the critic model long-term effects of the current prediction and incorporate this information into the gradient signal. Furthermore, to enable effective exploration in the inherently high-dimensional action space of instance masks, we learn a compact representation using a conditional variational auto-encoder. We show that our actor-critic model consistently provides accuracy benefits over the recurrent baseline on standard instance segmentation benchmarks.	de_DE
dc.language.iso	en	de_DE
dc.relation	IsDescribedBy;arXiv;1904.05126
dc.rights	Apache License 2.0
dc.rights.uri	https://www.apache.org/licenses/LICENSE-2.0
dc.subject	actor-critic	de_DE
dc.subject	reinforcement learning	de_DE
dc.subject	instance segmentation	de_DE
dc.subject.classification	409-05 Interaktive und intelligente Systeme, Bild- und Sprachverarbeitung, Computergraphik und Visualisierung	de_DE
dc.subject.ddc	004
dc.title	Actor-critic Instance Segmentation	de_DE
dc.type	Software	de_DE
tud.unit	TUDa