Recurrent Pixel Embedding for Instance Grouping

TL;DR

This paper presents a novel end-to-end trainable framework for pixel-level grouping tasks like instance segmentation, using hyper-spherical embeddings and a differentiable mean-shift clustering module, achieving improved results over existing methods.

Contribution

The introduction of a hyper-spherical embedding space and a recurrent, differentiable mean-shift clustering module for pixel grouping tasks is a novel approach.

Findings

Significant improvements in instance segmentation performance.

Effective grouping for boundary detection and semantic segmentation.

Theoretical analysis of embedding dimension and margin choices.

Abstract

We introduce a differentiable, end-to-end trainable framework for solving pixel-level grouping problems such as instance segmentation consisting of two novel components. First, we regress pixels into a hyper-spherical embedding space so that pixels from the same group have high cosine similarity while those from different groups have similarity below a specified margin. We analyze the choice of embedding dimension and margin, relating them to theoretical results on the problem of distributing points uniformly on the sphere. Second, to group instances, we utilize a variant of mean-shift clustering, implemented as a recurrent neural network parameterized by kernel bandwidth. This recurrent grouping module is differentiable, enjoys convergent dynamics and probabilistic interpretability. Backpropagating the group-weighted loss through this module allows learning to focus on only correcting embedding errors that won't be resolved during subsequent clustering. Our framework, while conceptually simple and theoretically abundant, is also practically effective and computationally efficient. We demonstrate substantial improvements over state-of-the-art instance segmentation for object proposal generation, as well as demonstrating the benefits of grouping loss for classification tasks such as boundary detection and semantic segmentation.