EVP: Enhanced Visual Perception using Inverse Multi-Attentive Feature Refinement and Regularized Image-Text Alignment
This work addresses the need for improved feature extraction and alignment in visual perception models, offering incremental enhancements over existing methods like VPD and ReLA for tasks such as depth estimation and segmentation.
The paper tackles the problem of enhancing visual perception for computer vision tasks by proposing the EVP architecture, which introduces an Inverse Multi-Attentive Feature Refinement module and a novel image-text alignment module, achieving state-of-the-art results such as an 11.8% RMSE improvement in single-image depth estimation on NYU Depth v2 and a 2.53 IoU improvement in referring segmentation on RefCOCO.
This work presents the network architecture EVP (Enhanced Visual Perception). EVP builds on the previous work VPD which paved the way to use the Stable Diffusion network for computer vision tasks. We propose two major enhancements. First, we develop the Inverse Multi-Attentive Feature Refinement (IMAFR) module which enhances feature learning capabilities by aggregating spatial information from higher pyramid levels. Second, we propose a novel image-text alignment module for improved feature extraction of the Stable Diffusion backbone. The resulting architecture is suitable for a wide variety of tasks and we demonstrate its performance in the context of single-image depth estimation with a specialized decoder using classification-based bins and referring segmentation with an off-the-shelf decoder. Comprehensive experiments conducted on established datasets show that EVP achieves state-of-the-art results in single-image depth estimation for indoor (NYU Depth v2, 11.8% RMSE improvement over VPD) and outdoor (KITTI) environments, as well as referring segmentation (RefCOCO, 2.53 IoU improvement over ReLA). The code and pre-trained models are publicly available at https://github.com/Lavreniuk/EVP.