Pedestrian detection is a key task in intelligent video surveillance systems which requires both fast inference and high detection accuracy.

A multi-scale deformable transformer encoder-based module that outperforms the state-of-the-art two-stage methods in detecting heavily occluded pedestrians ...

Apr 27, 2024 · These upsampled features contain high-level information, compensating for the lack of representation in lower features. [22] uses deformable ...

Deep neural networks targeting stereo disparity estimation have recently surpassed the performance of hand-crafted traditional models.

MULTI-SCALE DEFORMABLE TRANSFORMER ENCODER BASED SINGLE-STAGE PEDESTRIAN DETECTION. Jing Yuan, Tania Stathaki, Imperial College London, United Kingdom of ...

A multi-scale deformable transformer encoder-based module that outperforms the state-of-the-art two-stage methods in detecting heavily occluded pedestrians ...

Oct 31, 2022 · This paper improves single-stage pedestrian detectors with deformable vision transformers. The paper initially received mixed reviews, i.e. two ...

In this paper, we illustrate the effectiveness of the deformable vision transformer for single-stage pedestrian detection and propose a spatial and multi-scale ...

MS-DETR consists of two modality-specific backbones and Transformer encoders, followed by a multi-modal Transformer decoder, and the visible and thermal ...

This paper explores the research status of pedestrian detection in the past four years (2019–2022), focuses on analyzing the occlusion and scale problems of ...