paper-with-me

Papers

STeInFormer: Spatial-Temporal Interaction Transformer Architecture for Remote Sensing Change Detection

2024-12-23 · Xiaowen Ma, Zhenkai Wu, Mengting Ma, Mengjiao Zhao, Fan Yang, Zhenhong Du, Wei zhang

Convolutional neural networks and attention mechanisms have greatly benefited remote sensing change detection (RSCD) because of their outstanding discriminative ability. Existent RSCD methods often follow a paradigm of using a non-interactive Siamese neural network for multi-temporal feature extraction and change detection heads for feature fusion and change representation. However, this paradigm lacks the contemplation of the characteristics of RSCD in temporal and spatial dimensions, and causes the drawback on spatial-temporal interaction that hinders high-quality feature extraction. To address this problem, we present STeInFormer, a spatial-temporal interaction Transformer architecture for multi-temporal feature extraction, which is the first general backbone network specifically designed for RSCD. In addition, we propose a parameter-free multi-frequency token mixer to integrate frequency-domain features that provide spectral information for RSCD. Experimental results on three datasets validate the effectiveness of the proposed method, which can outperform the state-of-the-art methods and achieve the most satisfactory efficiency-accuracy trade-off. Code is available at https://github.com/xwmaxwma/rschange.

📄 PDF Abstract BibTeX arXiv:2412.17247

Code (1)

xwmaxwma/rschange 공식 구현 pytorch

Tasks

Change Detection

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Multi-Head Attention 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

Structure over Depth: A Single-Block Spatio-Temporal Transformer for Multi-Entity Reasoning

2026-07-25 · Narthana Sivalingam, Santhirarajah Sivasthigan, Buddhi Wijenayake, Roshan Godaliyadda 외 arxiv

Modeling multi-entity temporal data requires capturing dependencies across entities, time, and their interactions. Transformer-based approaches perform well but often rely on deep stacks of layers to learn these heteroge…

Group Activity Recognition

Beyond Appearance: Transformer-based Person Identification from Conversational Dynamics

2025-10-06 · Masoumeh Chapariniya, Teodora Vukovic, Sarah Ebling, Volker Dellwo arxiv

This paper investigates the performance of transformer-based architectures for person identification in natural, face-to-face conversation scenario. We implement and evaluate a two-stream framework that separately models…

Person IdentificationTransfer Learning

Boosting Urban Traffic Speed Prediction via Integrating Implicit Spatial Correlations

2022-12-25 · Dongkun Wang, Wei Fan, Pengyang Wang, Pengfei Wang 외

Urban traffic speed prediction aims to estimate the future traffic speed for improving the urban transportation services. Enormous efforts have been made on exploiting spatial correlations and temporal dependencies of tr…

Prediction

TubeDETR: Spatio-Temporal Video Grounding with Transformers

2022-03-30 · CVPR 2022 1 · Antoine Yang, Antoine Miech, Josef Sivic, Ivan Laptev 외

We consider the problem of localizing a spatio-temporal tube in a video corresponding to a given text query. This is a challenging task that requires the joint and efficient modeling of temporal, spatial and multi-modal …

DecoderLanguage-Based Temporal LocalizationNatural Language Visual Groundingobject-detection+6

A Gravity-informed Spatiotemporal Transformer for Human Activity Intensity Prediction

2025-06-16 · Yi Wang, Zhenghong Wang, Fan Zhang, Chengling Tang 외

Human activity intensity prediction is a crucial to many location-based services. Although tremendous progress has been made to model dynamic spatiotemporal patterns of human activity, most existing methods, including sp…