Distilling Object Detectors via Decoupled Features

Guo, Jianyuan; Han, Kai; Wang, Yunhe; Wu, Han; Chen, Xinghao; Xu, Chunjing; Xu, Chang

doi:10.1109/cvpr46437.2021.00219

Cited by 152 publications

(97 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This trend was initiated by Chen et al [5], which proposed to distill knowledge from a teacher detector to a student detector in both the backbone and head stages. Then, Wang et al [38] proposed to restrict the teacher-student feature imitation to regions around positive anchor boxes; Dai et al [8] produced general instances based on both the teacher's and student's outputs, and distilled feature-based, relation-based and response-based knowledge in these general instances; Guo et al [12] proposed to decouple the intermediate features and classification predictions of the positive and negative regions during knowledge distillation. All the aforementioned knowledge distillation methods require the student and the teacher to follow the same kind of detection framework, and thus typically transfer knowledge between models that only differ in terms of backbone, such as from a RetinaNet-ResNet152 to a RetinaNet-ResNet50.…”

Section: Related Workmentioning

confidence: 99%

“…To illustrate the generality of our approach, we also report the results of our distillation strategy used in conjunction with FKD [42], one of the current best detector-todetector distillation methods. Note that, while preparing this work, we also noticed the concurrent work of [12], whose DeFeat method also follows a detector-to-detector distillation approach, and thus could also be complemented with out strategy. Results.…”

Section: Comparison With Detector-to-detector Distillationmentioning

confidence: 99%

“…While early knowledge distillation techniques [18,34,35] focused on the task of image classification, several attempts have nonetheless been made for object detection. To this end, existing techniques [5,12,38] typically leverage the fact that object detection frameworks consist of three main stages depicted by Figure 1(a): A backbone to extract features; a neck to fuse the extracted features; and heads to predict classes and bounding boxes. Knowledge distillation is then achieved using a teacher with the same architecture as the student but a deeper and wider backbone, such as a Faster RCNN [33] with ResNet152 [14] teacher for a Faster RCNN with ResNet50 student, thus facilitating knowledge transfer at all three stages of the frameworks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Distilling Image Classifiers in Object Detectors

Guo

Álvarez

Salzmann

2021

Preprint

View full text Add to dashboard Cite

Knowledge distillation constitutes a simple yet effective way to improve the performance of a compact student network by exploiting the knowledge of a more powerful teacher. Nevertheless, the knowledge distillation literature remains limited to the scenario where the student and the teacher tackle the same task. Here, we investigate the problem of transferring knowledge not only across architectures but also across tasks. To this end, we study the case of object detection and, instead of following the standard detector-to-detector distillation approach, introduce a classifier-to-detector knowledge transfer framework. In particular, we propose strategies to exploit the classification teacher to improve both the detector's recognition accuracy and localization performance. Our experiments on several detectors with different backbones demonstrate the effectiveness of our approach, allowing us to outperform the state-of-the-art detector-to-detector distillation methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Comparison With Detector-to-detector Distillationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Distilling Image Classifiers in Object Detectors

Guo

Álvarez

Salzmann

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Among various research areas of AI, CV is a longstanding and fundamental field, which allows computers to derive meaningful information from digital images, videos, and other visual inputs. As a representative method in CV, CNN models have achieved the new SOTA performance on a wide range of tasks over the last few decades, e.g., Image Recognition [36], Object Detection [37]- [39], Image Segmentation [40]- [42], and Image Processing [43]- [46]. Image recognition involves analyzing images and identifying objects, actions, and other elements in order to draw conclusions.…”

Section: Computer Visionmentioning

confidence: 99%

“…In recent years, one-stage and two-stage object detectors [52]- [56] have achieved noticeable improvements. However, these methods rely on deep convolution operations to learn intensive features, resulting in a sharp increase in the cost of computing resources and an apparent decrease in detection speed [37]. Therefore, how to address these problems and enable real-time detection becomes an important line of research in object detection.…”

Section: Computer Visionmentioning

confidence: 99%

Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AI

Yao¹,

Zhang²,

Yang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Influenced by the great success of deep learning via cloud computing and the rapid development of edge chips, research in artificial intelligence (AI) has shifted to both of the computing paradigms, i.e., cloud computing and edge computing. In recent years, we have witnessed significant progress in developing more advanced AI models on cloud servers that surpass traditional deep learning models owing to model innovations (e.g., Transformers, Pretrained families), explosion of training data and soaring computing capabilities. However, edge computing, especially edge and cloud collaborative computing, are still in its infancy to announce their success due to the resource-constrained IoT scenarios with very limited algorithms deployed. In this survey, we conduct a systematic review for both cloud and edge AI. Specifically, we are the first to set up the collaborative learning mechanism for cloud and edge modeling with a thorough review of the architectures that enable such mechanism. We also discuss potentials and practical experiences of some on-going advanced edge AI topics including pretraining models, graph neural networks and reinforcement learning. Finally, we discuss the promising directions and challenges in this field.

show abstract

Distillation of human–object interaction contexts for action recognition

Almushyti

2022

Computer Animation & Virtual

View full text Add to dashboard Cite

Modeling spatial-temporal relations is imperative for recognizing human actions, especially when a human is interacting with objects, while multiple objects appear around the human differently over time. Most existing action recognition models focus on learning overall visual cues of a scene but disregard a holistic view of human-object relationships and interactions, that is, how a human interacts with respect to short-term task for completion and long-term goal. We therefore argue to improve human action recognition by exploiting both the local and global contexts of human-object interactions (HOIs). In this paper, we propose the Global-Local Interaction Distillation Network (GLIDN), learning human and object interactions through space and time via knowledge distillation for holistic HOI understanding. GLIDN encodes humans and objects into graph nodes and learns local and global relations via graph attention network. The local context graphs learn the relation between humans and objects at a frame level by capturing their co-occurrence at a specific time step.The global relation graph is constructed based on the video-level of human and object interactions, identifying their long-term relations throughout a video sequence. We also investigate how knowledge from these graphs can be distilled to their counterparts for improving HOI recognition. Finally, we evaluate our model by conducting comprehensive experiments on two datasets including Charades and CAD-120. Our method outperforms the baselines and counterpart approaches.

show abstract

Distilling Object Detectors via Decoupled Features

Cited by 152 publications

References 37 publications

Distilling Image Classifiers in Object Detectors

Distilling Image Classifiers in Object Detectors

Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AI

Distillation of human–object interaction contexts for action recognition

Contact Info

Product

Resources

About