IQDet: Instance-wise Quality Distribution Sampling for Object Detection

Ma, Yuchen; Liu, Songtao; Li, Zeming; Sun, Jian

doi:10.1109/cvpr46437.2021.00176

Cited by 47 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image object detection. In the era of deep learning, detection algorithms can be split into the two-stage [17,27,34,44,53,60] and the one-stage [14,28,31,32,35,38,39,41,51,61] frameworks. Some works, such as YOLO series [3,13,18,[41][42][43], adopt a bunch of training and accelerating tricks to achieve strong performance with real-time inference speed.…”

Section: Related Workmentioning

confidence: 99%

Real-time Object Detection for Streaming Perception

Yang¹,

Liu²,

Li³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Autonomous driving requires the model to perceive the environment and (re)act within a low latency for safety. While past works ignore the inevitable changes in the environment after processing, streaming perception is proposed to jointly evaluate the latency and accuracy into a single metric for video online perception. In this paper, instead of searching trade-offs between accuracy and speed like previous works, we point out that endowing real-time models with the ability to predict the future is the key to dealing with this problem. We build a simple and effective framework for streaming perception. It equips a novel Dual-Flow Perception module (DFP), which includes dynamic and static flows to capture the moving trend and basic detection feature for streaming prediction. Further, we introduce a Trend-Aware Loss (TAL) combined with a trend factor to generate adaptive weights for objects with different moving speeds. Our simple method achieves competitive performance on Argoverse-HD dataset and improves the AP by 4.9% compared to the strong baseline, validating its effectiveness. Our code will be made available at https: //github.com/yancie-yjr/StreamYOLO.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-time Object Detection for Streaming Perception

Yang¹,

Liu²,

Li³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, FCOS [40] and ATSS [47] introduce centerness branch for anchor-free detection. Other methods delve into sample assignment strategies [2,14,19,28,47,50].…”

Section: Related Workmentioning

confidence: 99%

R(Det)^2: Randomized Decision Routing for Object Detection

Li¹,

Wang²

2022

Preprint

View full text Add to dashboard Cite

In the paradigm of object detection, the decision head is an important part, which affects detection performance significantly. Yet how to design a high-performance decision head remains to be an open issue. In this paper, we propose a novel approach to combine decision trees and deep neural networks in an end-to-end learning manner for object detection. First, we disentangle the decision choices and prediction values by plugging soft decision trees into neural networks. To facilitate effective learning, we propose randomized decision routing with node selective and associative losses, which can boost the feature representative learning and network decision simultaneously. Second, we develop the decision head for object detection with narrow branches to generate the routing probabilities and masks, for the purpose of obtaining divergent decisions from different nodes. We name this approach as the randomized decision routing for object detection, abbreviated as R(Det) 2 . Experiments on MS-COCO dataset demonstrate that R(Det) 2 is effective to improve the detection performance. Equipped with existing detectors, it achieves 1.4 ∼ 3.6% AP improvement.

show abstract

“…The AB-OTA we proposed is an improvement based on YOLOv5 [6]'s label assignment rules [16,42,39,25,40], mainly referring to YOLOX's SimOTA [7]. YOLOv5 [6] is an anchor-based object detection model.…”

Section: Ab-otamentioning

confidence: 99%

YOLOSA: Object detection based on 2D local feature superimposed self-attention

Li¹,

Hua²

2022

Preprint

View full text Add to dashboard Cite

We analyzed the network structure of real-time object detection models and found that the features in the feature concatenation stage are very rich. Applying an attention module here can effectively improve the detection accuracy of the model. However, the commonly used attention module or self-attention module shows poor performance in detection accuracy and inference efficiency. Therefore, we propose a novel self-attention module, called 2D local feature superimposed self-attention, for the feature concatenation stage of the neck network. This self-attention module reflects global features through local features and local receptive fields. We also propose and optimize an efficient decoupled head and AB-OTA, and achieve SOTA results. Average precisions of 49.0% (66.2 FPS), 46.1% (80.6 FPS), and 39.1% (100 FPS) were obtained for large, medium, and small-scale models built using our proposed improvements. Our models exceeded YOLOv5 by 0.8% -3.1% in average precision.Preprint. Under review.

show abstract

IQDet: Instance-wise Quality Distribution Sampling for Object Detection

Cited by 47 publications

References 8 publications

Real-time Object Detection for Streaming Perception

Real-time Object Detection for Streaming Perception

R(Det)^2: Randomized Decision Routing for Object Detection

YOLOSA: Object detection based on 2D local feature superimposed self-attention

Contact Info

Product

Resources

About