Xiangyu Zhang

Xiangyu Zhang

Auto-generated profile, based on publicly available information

Xiangyu Zhang

Megvii Technology, Beijing, China

Research Focus & Metrics

Academic Career Length

23Years (2004-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

13 November 2022
Verified
Book Chapter
Closed access
PDF file
LCLiangyu Chen
Liangyu Chen
XCXiaojie Chu
Xiaojie Chu
XZXiangyu Zhang
Xiangyu Zhang
JSJian Sun
Jian Sun

30 October 2022
Verified
Book Chapter
Closed access
PDF file
YLYingfei Liu
Yingfei Liu
TWTiancai Wang
Tiancai Wang
XZXiangyu Zhang
Xiangyu Zhang
JSJian Sun
Jian Sun

6 October 2018
Verified
Book Chapter
Closed access
PDF file
ZZZhenli Zhang
Zhenli Zhang
XZXiangyu Zhang
Xiangyu Zhang
CPChao Peng
Chao Peng
XXXiangyang Xue
Xiangyang Xue
JSJian Sun
Jian Sun

5 October 2018
Verified
Book Chapter
Open access
PDF file
ZLZeming Li
Zeming Li
CPChao Peng
Chao Peng
Gang YuGang Yu
Gang Yu
XZXiangyu Zhang
Xiangyu Zhang
Yangdong DengYangdong Deng
Yangdong Deng
JSJian Sun
Jian Sun
Recent CNN based object detectors, either one-stage methods like YOLO, SSD, and RetinaNet, or two-stage detectors like Faster R-CNN, R-FCN and FPN, are usually trying to directly finetune from ImageNet pre-trained models designed for the task of image classification. However, there has been little work discussing the backbone feature extractor specifically designed for the task of object detection. More importantly, there are several differences between the tasks of image classification and object detection. (i) Recent object detectors like FPN and RetinaNet usually involve extra stages against the task of image classification to handle the objects with various scales. (ii) Object detection not only needs to recognize the category of the object instances but also spatially locate them. Large downsampling factors bring large valid receptive field, which is good for image classification, but compromises the object location ability. Due to the gap between the image classification and object detection, we propose DetNet in this paper, which is a novel backbone network specifically designed for object detection. Moreover, DetNet includes the extra stages against traditional backbone network for image classification, while maintains high spatial resolution in deeper layers. Without any bells and whistles, state-of-the-art results have been obtained for both object detection and instance segmentation on the MSCOCO benchmark based on our DetNet (4.8G FLOPs) backbone. Codes will be released ( https://github.com/zengarden/DetNet ).

15 May 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
CHChunrui Han
Chunrui Han
JYJinrong Yang
Jinrong Yang
JSJianjian Sun
Jianjian Sun
ZGZheng Ge
Zheng Ge
RDRunpei Dong
Runpei Dong
XZXiangyu Zhang
Xiangyu Zhang
Long-term temporal fusion is a crucial but often overlooked technique in camera-based Bird's-Eye-View (BEV) 3D perception. Existing methods are mostly in a parallel manner. While parallel fusion can benefit from long-term information, it suffers from increasing computational and memory overheads as the fusion window size grows. Alternatively, BEVFormer adopts a recurrent fusion pipeline so that history information can be efficiently integrated, yet it fails to benefit from longer temporal frames. In this letter, we explore an embarrassingly simple long-term recurrent fusion strategy built upon the LSS-based methods and find it already able to enjoy the merits from both sides, i.e., rich long-term information and efficient fusion pipeline. A temporal embedding module is further proposed to improve the model's robustness against occasionally missed frames in practical scenarios. We name this simple but effective fusing pipeline VideoBEV. Experimental results on the nuScenes benchmark show that VideoBEV obtains strong performance on various camera-based 3D perception tasks, including object detection ( 55.4% mAP and 62.9% NDS), segmentation ( 48.6% vehicle mIoU), tracking ( 54.8% AMOTA), and motion prediction ( 0.80 m minADE and 0.463 EPA).