CvT-ASSD: Convolutional vision-Transformer Based Attentive Single Shot MultiBox Detector

TL;DR

This paper introduces CvT-ASSD, a novel object detection model combining convolutional vision transformers with an efficient single shot detector, achieving good accuracy and efficiency on large-scale datasets.

Contribution

The paper proposes CvT-ASSD, integrating convolutional vision transformers with an attentive single shot detector to improve detection accuracy and computational efficiency.

Findings

Achieves competitive detection performance on PASCAL VOC and MS COCO datasets.

Reduces computational complexity compared to traditional transformer-based detectors.

Demonstrates effective balance between accuracy and efficiency.

Abstract

Due to the success of Bidirectional Encoder Representations from Transformers (BERT) in natural language process (NLP), the multi-head attention transformer has been more and more prevalent in computer-vision researches (CV). However, it still remains a challenge for researchers to put forward complex tasks such as vision detection and semantic segmentation. Although multiple Transformer-Based architectures like DETR and ViT-FRCNN have been proposed to complete object detection task, they inevitably decreases discrimination accuracy and brings down computational efficiency caused by the enormous learning parameters and heavy computational complexity incurred by the traditional self-attention operation. In order to alleviate these issues, we present a novel object detection architecture, named Convolutional vision Transformer Based Attentive Single Shot MultiBox Detector (CvT-ASSD), that built on the top of Convolutional vision Transormer (CvT) with the efficient Attentive Single Shot MultiBox Detector (ASSD). We provide comprehensive empirical evidence showing that our model CvT-ASSD can leads to good system efficiency and performance while being pretrained on large-scale detection datasets such as PASCAL VOC and MS COCO. Code has been released on public github repository at https://github.com/albert-jin/CvT-ASSD.