Feng Gao

Feng Gao

Auto-generated profile, based on publicly available information

Feng Gao

Meituan, Beijing, China

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
3 February 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
FGFeng Gao
Feng Gao
YCYeyun Cai
Yeyun Cai
FDFang Deng
Fang Deng
CYChengpu Yu
Chengpu Yu
JCJie Chen
Jie Chen
Most anchor-free methods perform object detection using dense recommendation, which assumes that one point can simultaneously conduct accurate category prediction and regression estimation. However, due to different task drivers, valid features for classification and regression may locate at distinct areas in the training phase. This problem is called feature misalignment. To solve it, we propose a new feature alignment method based on anchor-free object detector. Firstly, a global receptive field adaptor (G-RFA) is designed by incorporating the feature pyramid networks (FPN) with the global attention mechanism, and forward features are further fine-tuned with a deformable-subnet (De-Subnet) to remove the influence of redundant contextual information. Then, a new feature filter strategy with a misalignment score is proposed to guide the network to focus on sampling points with aligned features. In addition, we establish mutually independent multi-layer quality distributions to model the priori information of an object on different FPN levels. Equipped with our method, the classification and regression features are aligned, and the generated foreground weight map converges to the centers of classification and regression heatmaps. Experimental results show that without bells and whistles, our method achieves 49.3% AP on MS COCO test-dev under the default 2x training schedule, outperforming related methods. Besides, experiments on PASCAL VOC demonstrate the generalization ability of our method. Code is available at https://github.com/GFENGG/featurealign.

7 July 2021 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
FGFeng Gao
Feng Gao
FDFang Deng
Fang Deng
LLLinhan Li
Linhan Li
LZLele Zhang
Lele Zhang
JZJiaqi Zhu
Jiaqi Zhu
CYChengpu Yu
Chengpu Yu
Traditional monocular vision localization methods are usually suitable for short-range area and indoor relative positioning tasks. This paper presents MGG, a novel monocular global geolocation method for outdoor long-range targets. This method takes a single RGB image combined with necessary navigation parameters as input and outputs targets' GPS information under the Global Navigation Satellite System (GNSS). In MGG, we first design a camera pose correction method via pixel mapping to correct the pose of the camera. Then, we use anchor-based methods to improve the detection ability for long-range targets with small image regions. Next, the local monocular vision model (LMVM) with a local structure coefficient is proposed to establish an accurate 2D-to-3D mapping relationship. Subsequently, a soft correspondence constraint (SCC) is presented to solve the local structure coefficient, which can weaken the coupling degree between detection and localization. Finally, targets can be geolocated through optimization theory-based methods and a series of coordinate transformations. Furthermore, we demonstrate the importance of focal length on solving the error explosion problem in locating long-range targets with monocular vision. Extensive experiments on the challenging KITTI dataset as well as applications in outdoor environments with targets located at a long range of up to 150 meters show the superiority of our method.

4 May 2016 in Journal of Nuclear Science and Technology
Verified
Journal Article
Open access
HGHong Gao
Hong Gao
FGFeng Gao
Feng Gao
The mathematical models are developed to solve the non-dimensional transient flow rates in two loops and a reactor core under different power failures of reactor coolant pumps. Comparison of the experimental results of the Qinshan Nuclear Power Plant and the test results of the nuclear ship reactor in Japan under one pump power failure shows an excellent agreement. The non-dimensional flow rates are determined by the established non-dimensional parameters λ, ϵ, and α. Under the sequential power failure of two pumps, the non-dimensional flow rates are determined by the established λ, ϵ, α, and ΔT parameters. λ, ϵ, α, and ΔT are four important non-dimensional parameters in the prediction of flow transients. λ indicates the resistance coefficient ratio of the single loop to the reactor core, ϵ indicates the fluid inertia ratio of the reactor core to the single loop, α indicates the ratio of the initial kinetic energy of the single loop coolant fluid to the effective initial kinetic energy of the reactor coolant pump, and ΔT means the non-dimensional time interval of the sequential power failure of two pumps. The effects of λ, ϵ, α, and ΔT on the non-dimensional flow rates and the temperature change are investigated.

27 July 2025
Verified
Conference Paper
Open access
ZKZhe Kong
Zhe Kong
LLLe Li
Le Li
YZYong Zhang
Yong Zhang
FGFeng Gao
Feng Gao
SYShaoshu Yang
Shaoshu Yang
WLWenhan Luo
Wenhan Luo

19 July 2026
Verified
Conference Paper
Open access
YPYouxin Pang
Youxin Pang
JLJiajun Liu
Jiajun Liu
LTLingfeng Tan
Lingfeng Tan
YZYong Zhang
Yong Zhang
FGFeng Gao
Feng Gao
YLYebin Liu
Yebin Liu