Multimodal Object Detection by Channel Switching and Spatial Attention
The DOI matches its Crossref record; 32 of the 32 references it cites were found in a registry or an index.
Abstract
Multimodal object detection has attracted great attention in recent years since the information specific to different modalities can complement each other and effectively improve the accuracy and stability of the detection model. However, compared to processing the inputs from a single modality, fusing information from multiple modalities can significantly increase the computational complexity of the model, thus impairing its efficiency. Therefore the multi-modal fusion module needs to be carefully designed to enhance the performance of the detection model while keeping the computational consumption low. In this paper, we propose a novel lightweight fusion module that can efficiently fuse the inputs from different modalities using channel switching and spatial attention (CSSA). The effectiveness and generalizability of the module are tested using two public multimodal datasets LLVIP and FLIR, both of which comprise paired infrared (IR) and visible (RGB) images. The experiments demonstrate that the proposed CSSA module can substantially improve the accuracy of multimodal object detection without consuming excessive computing resources.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 2×
“Since the objective of multimodal detection is the same as unimodal detection, many related studies are based on the traditional RGB detection models such as Reti-naNet [17], and YOLO [21].”
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Zhang, X., Zhou, X., Lin, M., & Sun, J. (2017) · arXiv · doi:10.48550/arxiv.1707.01083 · Cited by 9,982
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Zhang, X., Zhou, X., Lin, M., & Sun, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00716 · Cited by 9,982
ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
Wang, Q., Wu, B., Zhu, P., Li, P., Zuo, W., & Hu, Q. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.01155 · Cited by 9,641 · cited 5×
“ECA can perceive local cross-channel interaction efficiently and can be described as: ωm = σ (f (GAP (Xm))) (1) where GAP(X) = 1 HW H−1∑ i=0 W−1∑ j=0 Xij (2) where Xm ∈ RH×W×C is the feature maps obtained from modality m (a height×width×channel tensor), and f(X) is a 1D convolution layer.”
MethodInfluentialLLVIP: A Visible-infrared Paired Dataset for Low-light Vision
Jia, X., Zhu, C., Li, M., Tang, W., & Zhou, W. (2021) · 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) · doi:10.1109/iccvw54120.2021.00389 · Cited by 896 · cited 7×
“The performance of the proposed model and the baselines mentioned above are evaluated on the FLIR [30] and LLVIP [11] datasets.”
BackgroundMethodInfluentialDrone-Based RGB-Infrared Cross-Modality Vehicle Detection Via Uncertainty-Aware Learning
Sun, Y., Cao, B., Zhu, P., & Hu, Q. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3168279 · Cited by 693 · cited 2×
“The authors of UACMDet [23] report in their paper that the time required by UA-CMDet to process a single input is 370ms, which is an unacceptable inference time for many time-critical tasks.”
BackgroundMultispectral Deep Neural Networks for Pedestrian Detection
Liu, J., Zhang, S., Wang, S., & Metaxas, D. (2016) · Procedings of the British Machine Vision Conference 2016 · doi:10.5244/c.30.73 · Cited by 543
Illumination-aware faster R-CNN for robust multispectral pedestrian detection
Li, C., Song, D., Tong, R., & Tang, M. (2019) · Pattern Recognition · doi:10.1016/j.patcog.2018.08.005 · Cited by 533
TransFuser: Imitation With Transformer-Based Sensor Fusion for Autonomous Driving
Chitta, K., Prakash, A., Jaeger, B., Yu, Z., Renz, K., & Geiger, A. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3200245 · Cited by 505
Guan, D., Cao, Y., Yang, J., Cao, Y., & Yang, M. Y. (2019) · Information Fusion · doi:10.1016/j.inffus.2018.11.017 · Cited by 342
Self-citationMultimodal end-to-end autonomous driving
Xiao, Y., Codevilla, F., Gurram, A., Urfalioglu, O., & Lopez, A. M. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.3013234 · Cited by 319
Cross-modality attentive feature fusion for object detection in multispectral remote sensing imagery
Qingyun, F., & Zhaokui, W. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108786 · Cited by 303
Multispectral Fusion for Object Detection with Cyclic Fuse-and-Refine Blocks
Zhang, H., Fromont, E., Lefevre, S., & Avignon, B. (2020) · 2020 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip40778.2020.9191080 · Cited by 297 · cited 9×
“The performance of the proposed model and the baselines mentioned above are evaluated on the FLIR [30] and LLVIP [11] datasets.”
MethodInfluentialGuided Attentive Feature Fusion for Multispectral Pedestrian Detection
ZHANG, H., FROMONT, E., LEFEVRE, S., & AVIGNON, B. (2021) · 2021 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv48630.2021.00012 · Cited by 290
Multispectral Pedestrian Detection via Simultaneous Detection and SegmentationPreprint
Li, C., Song, D., Tong, R., & Tang, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.04818 · Cited by 273
Weakly Aligned Cross-Modal Learning for Multispectral Pedestrian Detection
Zhang, L., Zhu, X., Chen, X., Yang, X., Lei, Z., & Liu, Z. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00523 · Cited by 271 · cited 1×
“In late-fusion (decision fusion), the inputs of the two modalities are fed separately into two unimodal object detection models to generate bounding boxes, and the predicted bounding boxes are fused using statistical methods [3, 13, 31].”
MethodSelf-citationMultispectral Pedestrian Detection using Deep Fusion Convolutional Neural Networks.
Wagner, J., Fischer, V., Herman, M., & Behnke, S. (2016) · The European Symposium on Artificial Neural Networks · Cited by 244
Learning Cross-Modal Deep Representations for Robust Pedestrian Detection
Xu, D., Ouyang, W., Ricci, E., Wang, X., & Sebe, N. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.451 · Cited by 237 · cited 1×
“Multimodal object detection can be divided into three categories in light of the fusion strategy, including earlyfusion, late-fusion, and mid-fusion [5, 13, 19, 24, 29].”
BackgroundMultimodal Object Detection via Probabilistic Ensembling
Chen, Y.-T., Shi, J., Ye, Z., Mertz, C., Ramanan, D., & Kong, S. (2022) · Lecture Notes in Computer Science · doi:10.1007/978-3-031-20077-9_9 · Cited by 225 · cited 7×
“For example, [3] uses Probabilistic Ensembling, which allows the model to cope with unaligned data.”
MethodInfluentialDeep Multimodal Fusion by Channel Exchanging
Wang, Y., Huang, W., Sun, F., Xu, T., Rong, Y., & Huang, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.05005 · Cited by 110 · cited 1×
“To perform channel attention, channel switching is applied in our module, as it is efficient and effective for feature interaction across modalities [26, 32].”
MethodMultispectral Image Fusion and Colorization
Blasch, E., Zheng, Y., & Liu, Z. (2018) · SPIE eBooks · doi:10.1117/3.2316455 · Cited by 110
Self-citationChoi, H., Kim, S., Park, K., & Sohn, K. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899703 · Cited by 101 · cited 1×
“Multimodal object detection can be divided into three categories in light of the fusion strategy, including earlyfusion, late-fusion, and mid-fusion [5, 13, 19, 24, 29].”
BackgroundDeep Visible and Thermal Image Fusion for Enhanced Pedestrian Visibility
Shopovska, I., Jovanov, L., & Philips, W. R. (2019) · Sensors · doi:10.3390/s19173727 · Cited by 90 · cited 1×
“has become a typical way to improve the performance of the model [2, 7, 22, 28].”
BackgroundTask-Conditioned Domain Adaptation for Pedestrian Detection in Thermal Imagery
Kieu, M., Bagdanov, A. D., Bertini, M., & Bimbo, A. D. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58542-6_33 · Cited by 85 · cited 1×
“Multimodal object detection can be divided into three categories in light of the fusion strategy, including earlyfusion, late-fusion, and mid-fusion [5, 13, 19, 24, 29].”
BackgroundContext-Aware Fusion of RGB and Thermal Imagery for Traffic Monitoring
Alldieck, T., Bahnsen, C., & Moeslund, T. B. (2016) · Sensors · doi:10.3390/s16111947 · Cited by 77
Multi-Channel CNN-based Object Detection for Enhanced Situation AwarenessPreprint
Liu, S., & Liu, Z. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.00075 · Cited by 60 · cited 1×
“However, uni-modal object detection, which is restricted by environmental factors, is sometimes insufficient for all realistic scenarios [18, 33].”
Self-citationLIDAR and Monocular Camera Fusion: On-road Depth Completion for Autonomous Driving
Fu, C., Mertz, C., & Dolan, J. M. (2019) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2019.8917201 · Cited by 56
Metadata
- Authors
- Cao, Y., Bin, J., Hamari, J., Blasch, E., & Liu, Z.
- Venue
- 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
- Publisher
- IEEE
- Published
- 2023-06
- Pages
- 403-411
- OpenAlex
- W4385801610
- Cited by
- 190
- References
- 32
- FWCI
- 21.60
- Type
- conference paper
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.