Bayesian Semantic Instance Segmentation in Open Set World

By Trung T. Pham, G. VijayKumarB., Thanh-Toan Do, Gustavo Carneiro, Ian David Reid. 2018

35 citations recorded by the source corpus.

This paper addresses the semantic instance segmentation task in the open-set conditions, where input images can contain known and unknown object classes. The training process of existing semantic instance segmentation methods requires annotation masks for all object instances, which is expensive to acquire or even infeasible in some realistic scenarios, where the number of categories may increase boundlessly. In this paper, we present a novel open-set semantic instance segmentation approach capable of segmenting all known and unknown object classes in images, based on the output of an object detector trained on known object classes. We formulate the problem using a Bayesian framework, where the posterior distribution is approximated with a simulated annealing optimization equipped with an efficient image partition sampler. We show empirically that our method is competitive with state-of-the-art supervised methods on known classes, but also performs well on unknown classes when compared with unsupervised methods.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Probabilistic & Bayesian Methods

Citations in the corpus

Learning Open-World Object Proposals without Learning to Classify (2021, background): “stereo images [37], or formulated with boundary detection as a prior [40] assuming objects share strong edges.”
Recent Advances in Open Set Recognition: A Survey (2018, background): “For example, [155]–[157] have introduced this scenario into domain adaptation, while [158] introduced it to the semantic instance segmentation task.”
Opening up Open World Tracking (2021, methodology): “Open-world instance segmentation was tackled in [76] by combining data-driven [34] and model-based segmentation methods [28, 80].”
ShapeMask: Learning to Segment Novel Objects by Refining Shape Priors (2019, background): “[37] studies open-set instance segmentation by”
Uncertainty-Aware Learning for Zero-Shot Semantic Segmentation (2020, methodology): “To achieve effective zero-shot semantic segmentation (ZSS), existing efforts have been made [2, 28, 31, 44, 55, 62] by treating each pixel as an independent classiﬁcation problem.”