Lite DETR : An Interleaved Multi-Scale Encoder for Efficient DETR
Open accessAbstract
Recent DEtection TRansformer-based (DETR) models have obtained remarkable performance. Its success cannot be achieved without the re-introduction of multi-scale feature fusion in the encoder. However, the excessively increased tokens in multi-scale features, especially for about 75% of low-level features, are quite computationally inefficient, which hinders real applications of DETR models. In this paper, we present Lite DETR, a simple yet efficient end-to-end object detection framework that can effectively reduce the GFLOPs of the detection head by 60% while keeping 99% of the original performance. Specifically, we design an efficient encoder block to update high-level features (corresponding to small-resolution feature maps) and low-level features (corresponding to large-resolution feature maps) in an interleaved way. In addition, to better fuse cross-scale features, we develop a key-aware deformable attention to predict more reliable attention weights. Comprehensive experiments validate the effectiveness and efficiency of the proposed Lite DETR, and the efficient encoder strategy can generalize well across existing DETR-based models. The code will be available in https://github.com/IDEA-Research/Lite-DETR.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
DAB-DETR and DN-DETR improve performance by introducing the iterative refinement scheme and denoising training.
“DAB-DETR [23] and DN-DETR [17] further improve performance by introducing the iterative refinement scheme and denoising training.”
From DETRs Beat YOLOs on Real-time Object Detection (2024) · cites this paper as [17]
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 · see citing works
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00986 · Cited by 35,534 · see citing works
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 · see citing works
End-to-End Object Detection with Transformers
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58452-8_13 · Cited by 18,285 · see citing works
YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
Wang, C.-Y., Bochkovskiy, A., & Liao, H.-Y. M. (2022) · arXiv · doi:10.48550/arxiv.2207.02696 · Cited by 13,205 · see citing works
YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors
Wang, C.-Y., Bochkovskiy, A., & Liao, H.-Y. M. (2023) · 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52729.2023.00721 · Cited by 13,198 · see citing works
EfficientDet: Scalable and Efficient Object Detection
Tan, M., Pang, R., & Le, Q. V. (2019) · arXiv · Cited by 9,841 · cited 5× · see citing works
“All models except EfficientDet [29] and YOLO series [12, 30] use ResNet-50 and Swin-Tiny as backbones.”
BackgroundMethodInfluentialDeformable DETR: Deformable Transformers for End-to-End Object Detection
Zhu, X., Su, W., Lu, L., Li, B., Wang, X., & Dai, J. (2020) · arXiv · doi:10.48550/arxiv.2010.04159 · Cited by 1,837 · cited 10× · see citing works
“Recently, DEtection TRansformer [1] (DETR) introduces Transformers into object detection, and DETR-like models have achieved promising performance on many fundamental vision tasks, such as object detection [13, 36, 37], instance segmentation [5, 6, 14], and pose estimation [26, 28].”
BackgroundMethodInfluentialDINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection
Zhang, H., Li, F., Liu, S., Zhang, L., Su, H., Zhu, J., Ni, L. M., & Shum, H.-Y. (2022) · arXiv · Cited by 1,154 · cited 10× · see citing works
“Improving Decoder Design of DETR: Recently, DETR-based detectors have seen more rapid progress [13, 18, 21, 33, 36] compared to classical detectors [2,24].”
BackgroundMethodInfluentialSelf-citationDN-DETR: Accelerate DETR Training by Introducing Query DeNoising
Li, F., Zhang, H., Liu, S., Guo, J., Ni, L. M., & Zhang, L. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01325 · Cited by 1,102 · see citing works
Self-citationConditional DETR for Fast Training Convergence
Meng, D., Chen, X., Fan, Z., Zeng, G., Li, H., Yuan, Y., Sun, L., & Wang, J. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00363 · Cited by 1,055 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
DETRs Beat YOLOs on Real-time Object Detection
Zhao, Y., Lv, W., Xu, S., Wei, J., Wang, G., Dang, Q., Liu, Y., & Chen, J. (2024) · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52733.2024.01605 · Cited by 5,020 · see citing works
“Lite DETR [17] enhances the efficiency of encoder by reducing the update frequency of low-level features in an interleaved way.”
Background
- Authors
- Li, F., Zeng, A., Liu, S., Zhang, H., Li, H., Zhang, L., & Ni, L. M.
- Affiliations
- Hong Kong University of Science and Technology (HK)
- International Digital Economy Academy (IDEA)
- Tsinghua University (CN)
- Venue
- 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2023-06
- Pages
- 18558-18567
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4386076079
- Cited by
- 154
- References
- 11
- Citations by year
- 2026: 47 · 2025: 56 · 2024: 40 · 2023: 11
- FWCI
- 16.47
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- DETR, object detection, multi-scale feature fusion, computational efficiency, low-level features, high-level features, cross-scale feature fusion, attention weights
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
