Segment-Phrase Table for Semantic Segmentation, Visual Entailment and Paraphrasing

TL;DR

This paper introduces the Segment-Phrase Table, a large resource linking textual phrases to image segments, which enhances semantic segmentation, visual entailment, and paraphrasing through minimal supervision and rich bimodal data.

Contribution

The paper presents a novel high-quality segment-phrase table that improves semantic segmentation and enables new applications like visual entailment and paraphrasing with minimal supervision.

Findings

Achieved state-of-the-art segmentation results on benchmark datasets.

Enabled richer semantic understanding and reasoning of textual phrases.

Demonstrated utility in visual entailment and paraphrasing tasks.

Abstract

We introduce Segment-Phrase Table (SPT), a large collection of bijective associations between textual phrases and their corresponding segmentations. Leveraging recent progress in object recognition and natural language semantics, we show how we can successfully build a high-quality segment-phrase table using minimal human supervision. More importantly, we demonstrate the unique value unleashed by this rich bimodal resource, for both vision as well as natural language understanding. First, we show that fine-grained textual labels facilitate contextual reasoning that helps in satisfying semantic constraints across image segments. This feature enables us to achieve state-of-the-art segmentation results on benchmark datasets. Next, we show that the association of high-quality segmentations to textual phrases aids in richer semantic understanding and reasoning of these textual phrases. Leveraging this feature, we motivate the problem of visual entailment and visual paraphrasing, and demonstrate its utility on a large dataset.