Semantically Stable Image Composition Analysis via Saliency and Gradient Vector Flow Fusion

TL;DR

This paper introduces VFCNet, a novel model that fuses saliency and edge information into a gradient vector flow field to assess photographic composition, achieving state-of-the-art results on the PICD benchmark.

Contribution

VFCNet is a new approach that combines saliency and edge cues into a gradient vector flow for robust composition analysis, outperforming previous methods.

Findings

VFCNet achieves CDA-1 of 0.683 and CDA-2 of 0.629 on PICD.

The model improves performance by over 33% and 36% compared to prior best methods.

A simple classifier on self-supervised features surpasses specialized composition models.

Abstract

The reliable computational assessment of photographic composition requires features that are discriminative of spatial layout yet robust to semantic content. This paper proposes a low-level representation grounded in the assumption that composition can be understood as the flow of visual attention across geometric structure. We introduce VFCNet, which fuses saliency and edge information into a gradient vector flow (GVF) field. The model computes dual-stream GVF representations, integrates them via attention, and extracts multi-scale flow features with a DINOv3 backbone. VFCNet achieves state-of-the-art performance on the PICD benchmark (CDA-1: 0.683, CDA-2: 0.629), improving by 33.1\% and 36.1\% over the previous best method. We also show that a simple classifier on self-supervised DINOv3 features substantially outperforms more sophisticated, composition-specialized models. Code is available at https://github.com/ADadras/VFCNet