TOOD: Task-aligned One-stage Object Detection

By Chengjian Feng, Yujie Zhong, Yu Gao, Matthew R. Scott, Weilin Huang. IEEE International Conference on Computer Vision, 2021

1,333 citations recorded by the source corpus.

One-stage object detection is commonly implemented by optimizing two sub-tasks: object classification and localization, using heads with two parallel branches, which might lead to a certain level of spatial misalignment in predictions between the two tasks. In this work, we propose a Task-aligned One-stage Object Detection (TOOD) that explicitly aligns the two tasks in a learning-based manner. First, we design a novel Task-aligned Head (T-Head) which offers a better balance between learning task-interactive and task-specific features, as well as a greater flexibility to learn the alignment via a task-aligned predictor. Second, we propose Task Alignment Learning (TAL) to explicitly pull closer (or even unify) the optimal anchors for the two tasks during training via a designed sample assignment scheme and a task-aligned loss. Extensive experiments are conducted on MS-COCO, where TOOD achieves a 51.1 AP at single-model single-scale testing. This surpasses the recent one-stage detectors by a large margin, such as ATSS [30] (47.7 AP), GFL [14] (48.2 AP), and PAA [9] (49.0 AP), with fewer parameters and FLOPs. Qualitative results also demonstrate the effectiveness of TOOD for better aligning the tasks of object classification and localization. Code is available at https://github.com/fcjian/TOOD.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7438524bf00d7c5a22cb8799797f57c3a794b220

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Towards Large-Scale Small Object Detection: Survey and Benchmarks (2022, background): “Not surprisingly, this powerful mechanism has been extensively investigated in the previous literature [96], [97], [98], [99], [100] and shows great potential in many vision ﬁelds [5], [9], [101], [102].”
YOLO-World: Real-Time Open-Vocabulary Object Detection (2024, methodology): “We follow [20] and leverage task-aligned label assignment [10] to match the predictions with ground-truth annotations and assign each positive prediction with a text index as the classification label.”
YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors (2022, methodology): “However, in recent years, if we take object detection as an example, researchers often use the quality and distribution of prediction output by the network, and then consider together with the ground truth to use some calculation and optimization methods to generate a reliable soft label [61, 8, 36,”
YOLOv10: Real-Time End-to-End Object Detection (2024, methodology): “We can then leverage the normalized metric [15] to obtain the classification target for task alignment learning [21, 15, 65, 29, 70] with our holistic efficiency-accuracy driven model design, showing a 2.0% AP improvement with only 0.05ms latency overhead.”
A Comprehensive Review of YOLO: From YOLOv1 to YOLOv8 and Beyond (2023, methodology): “Label assignment using the Task alignment learning approach introduced in TOOD [88].”
YOLO-based Object Detection Models: A Review and its Applications (2024, positive): “VFL is derived from focal loss by maintaining the negative and positive data at different degrees of importance for label assignment using a learning approach called task alignment introduced in TOOD [85].”
YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications (2022, methodology): “• Label Assignment: We evaluate the recent progress of label assignment strategies [5, 7, 18, 48, 51] on YOLOv6 through numerous experiments, and the results indicate that TAL [5] is more effective and training-friendly.”
PP-YOLOE: An evolved version of YOLO (2022, methodology): “2, we use ESE to replace the layer attention in TOOD, simplify the alignment of classification branches to shortcut, and replace the alignment of regression branches with distribution focal loss (DFL) layer[16].”
YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information (2024, background): “In addition, some researchers have tried to develop more general objective functions, such as loss function [5, 45, 46, 50, 77, 78], label assignment [10, 12, 33, 67, 79] and auxiliary supervision [18, 20, 24, 28, 29, 51, 54, 68, 76].”
A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS (2023, methodology): “Label assignment using the Task alignment learning approach introduced in TOOD [100].”
YOLOv12: Attention-Centric Real-Time Object Detectors (2025, neutral): “Although improvements in YOLO have focused on areas such as loss functions [8, 35, 43, 44, 48, 67, 68], label assignment [22, 23, 34, 59, 69], network architecture design has remained a critical research priority [24, 28, 32, 57, 58].”
RTMDet: An Empirical Study of Designing Real-Time Object Detectors (2022, methodology): “To train the one-stage object detector, the dense predictions from each scale will be matched with ground truth bounding boxes through different label assignment strategies [19, 47, 70].”
UAV-YOLOv8: A Small-Object-Detection Model Based on Improved YOLOv8 for UAV Aerial Photography Scenarios (2023, methodology): “It also uses the Task-Aligned Assigner [36] to dynamically assign samples, which improves the detection accuracy and robustness of the model.”
DAMO-YOLO : A Report on Real-Time Object Detection Design (2022, methodology): “In addition, the dynamic label assignment, such as OTA [8] and TOOD [7], is widely acclaimed and achieves signiﬁcant improvement compared to the static label assignment [43].”