DynaMITe: Dynamic Query Bootstrapping for Multi-object Interactive Segmentation Transformer

IEEE/CVF International Conference on Computer Vision (ICCV), pp. 1043–1052

Abstract

Most state-of-the-art instance segmentation methods rely on large amounts of pixel-precise ground-truth annotations for training, which are expensive to create. Interactive segmentation networks help generate such annotations based on an image and the corresponding user interactions such as clicks. Existing methods for this task can only process a single instance at a time and each user interaction requires a full forward pass through the entire deep network. We introduce a more efficient approach, called DynaMITe, in which we represent user interactions as spatio-temporal queries to a Transformer decoder with a potential to segment multiple object instances in a single iteration. Our architecture also alleviates any need to re-compute image features during refinement, and requires fewer interactions for segmenting multiple instances in a single image when compared to other methods. DynaMITe achieves state-of-the-art results on multiple existing interactive segmentation benchmarks, and also on the new multi-instance benchmark that we propose in this paper.

Authors 3

1. Sabarinath Mahadevan Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Germany

   RWTH Aachen University, Germany

2. Alexander Hermans Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Germany

   RWTH Aachen University, Germany

3. Bastian Leibe Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Germany

   RWTH Aachen University, Germany