Jiefeng Li
YOU?
Author Swipe
View article: Distinct Microphysical Characteristics of Precipitating and Non‐Precipitating Parts of a Stratus Cloud From In Situ Aircraft Observations
Distinct Microphysical Characteristics of Precipitating and Non‐Precipitating Parts of a Stratus Cloud From In Situ Aircraft Observations Open
Cloud microphysical properties play a crucial role in stratus cloud behavior, with notable differences between precipitating and non‐precipitating parts. This study presents aircraft‐based observations on 8 May 2020 over a coastal region i…
View article: Training and Generalization of a Fault Classification AI Model for Lift Door System Using Data Augmentation
Training and Generalization of a Fault Classification AI Model for Lift Door System Using Data Augmentation Open
Data augmentation is an effective method in machine learning to address data scarcity. When the cost of acquiring specific state data in professional fields are high, data augmentation can expand the limited datasets and improve model perf…
View article: Weakened snow and ice melting by enhanced cloud short-wave cooling effect in the Arctic
Weakened snow and ice melting by enhanced cloud short-wave cooling effect in the Arctic Open
Clouds and surface albedo significantly affect the energy balance on Earth. With the snow/ice melting in the Arctic, an understanding of how this impacts short-wave cloud radiative effects (CRE) remains a critical, yet poorly understood, q…
View article: 2000~2022年北极夏季云的变化趋势及其潜在成因
2000~2022年北极夏季云的变化趋势及其潜在成因 Open
View article: Correction: Immunotoxicity assessment of CdSe/ZnS quantum dots in macrophages, lymphocytes and BALB/c mice
Correction: Immunotoxicity assessment of CdSe/ZnS quantum dots in macrophages, lymphocytes and BALB/c mice Open
View article: “北极放大”现象驱动机制的研究进展
“北极放大”现象驱动机制的研究进展 Open
View article: Surface albedo regulates aerosol direct climate effect
Surface albedo regulates aerosol direct climate effect Open
Aerosols and Surface Albedo (SA) are critical in balancing Earth's energy budget. With the changes of surface types and corresponding SA in recent years, an intriguing yet unresolved question emerges: how does Aerosol Direct Radiative Effe…
View article: Effects of herbal dregs supplementation of Salvia miltiorrhiza and Isatidis Radix residues improved production performance and gut microbiota abundance in late-phase laying hens
Effects of herbal dregs supplementation of Salvia miltiorrhiza and Isatidis Radix residues improved production performance and gut microbiota abundance in late-phase laying hens Open
The present study was designed to evaluate the effect of a mixture of Chinese medicinal residues (CMRs) consisting of Salvia miltiorrhiza residues (SMR) and Isatidis Radix residues (IRR) on productive performance, egg quality, serum lipid …
View article: Quantitative lipidomics reveals the changes of lipids and antioxidant capacity in egg yolk from laying hens with fatty liver hemorrhagic syndrome
Quantitative lipidomics reveals the changes of lipids and antioxidant capacity in egg yolk from laying hens with fatty liver hemorrhagic syndrome Open
In laying hens, fatty liver hemorrhagic syndrome (FLHS) is a common metabolic disorder, which can affect egg production and nutritional value. However, the impact of FLHS on the lipid content in egg yolks was not clear. In this study, FLHS…
View article: ShapeBoost: Boosting Human Shape Estimation with Part-Based Parameterization and Clothing-Preserving Augmentation
ShapeBoost: Boosting Human Shape Estimation with Part-Based Parameterization and Clothing-Preserving Augmentation Open
Accurate human shape recovery from a monocular RGB image is a challenging task because humans come in different shapes and sizes and wear different clothes. In this paper, we propose ShapeBoost, a new human shape recovery framework that ac…
View article: ShapeBoost: Boosting Human Shape Estimation with Part-Based Parameterization and Clothing-Preserving Augmentation
ShapeBoost: Boosting Human Shape Estimation with Part-Based Parameterization and Clothing-Preserving Augmentation Open
Accurate human shape recovery from a monocular RGB image is a challenging task because humans come in different shapes and sizes and wear different clothes. In this paper, we propose ShapeBoost, a new human shape recovery framework that ac…
View article: Denoising Vision Transformers
Denoising Vision Transformers Open
We study a crucial yet often overlooked issue inherent to Vision Transformers (ViTs): feature maps of these models exhibit grid-like artifacts, which hurt the performance of ViTs in downstream dense prediction tasks such as semantic segmen…
View article: Characterization of atmospheric microplastics in Hangzhou, a megacity of the Yangtze river delta, China
Characterization of atmospheric microplastics in Hangzhou, a megacity of the Yangtze river delta, China Open
Atmospheric microplastics in Hangzhou were predominantly fibers, with tires (27.0%) and PET (19.7%) as the dominant types. The annual MP dry deposition flux in the Hangzhou urban area was estimated to be 16.9 ± 2.9 tons.
View article: NIKI: Neural Inverse Kinematics with Invertible Neural Networks for 3D Human Pose and Shape Estimation
NIKI: Neural Inverse Kinematics with Invertible Neural Networks for 3D Human Pose and Shape Estimation Open
With the progress of 3D human pose and shape estimation, state-of-the-art methods can either be robust to occlusions or obtain pixel-aligned accuracy in non-occlusion cases. However, they cannot obtain robustness and mesh-image alignment a…
View article: HybrIK-X: Hybrid Analytical-Neural Inverse Kinematics for Whole-body Mesh Recovery
HybrIK-X: Hybrid Analytical-Neural Inverse Kinematics for Whole-body Mesh Recovery Open
Recovering whole-body mesh by inferring the abstract pose and shape parameters from visual content can obtain 3D bodies with realistic structures. However, the inferring process is highly non-linear and suffers from image-mesh misalignment…
View article: Relation analysis of bacterial community in soils of coal mines with potential ecological risk from heavy metals
Relation analysis of bacterial community in soils of coal mines with potential ecological risk from heavy metals Open
Coal consumption increases consistently, and heavy metal contamination in soils surrounding coal mines has been a severe environmental issue. Investigation of the relations of the bacterial community in soils of coal mines with potential e…
View article: AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time
AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time Open
Accurate whole-body multi-person pose estimation and tracking is an important yet challenging topic in computer vision. To capture the subtle actions of humans for complex behavior analysis, whole-body pose estimation including the face, b…
View article: Constructing Balance from Imbalance for Long-tailed Image Recognition
Constructing Balance from Imbalance for Long-tailed Image Recognition Open
Long-tailed image recognition presents massive challenges to deep learning systems since the imbalance between majority (head) classes and minority (tail) classes severely skews the data-driven deep neural networks. Previous methods tackle…
View article: Unsupervised Representation for Semantic Segmentation by Implicit Cycle-Attention Contrastive Learning
Unsupervised Representation for Semantic Segmentation by Implicit Cycle-Attention Contrastive Learning Open
We study the unsupervised representation learning for the semantic segmentation task. Different from previous works that aim at providing unsupervised pre-trained backbones for segmentation models which need further supervised fine-tune, h…
View article: Correlation Field for Boosting 3D Object Detection in Structured Scenes
Correlation Field for Boosting 3D Object Detection in Structured Scenes Open
Data augmentation is an efficient way to elevate 3D object detection performance. In this paper, we propose a simple but effective online crop-and-paste data augmentation pipeline for structured 3D point cloud scenes, named CorrelaBoost. O…
View article: Label-Free Quantitative Proteomic Analysis Provides Insight Into the Differences Between Slow-Twitch Muscle and Fast-Twitch Muscle of Pseudocaranx dentex
Label-Free Quantitative Proteomic Analysis Provides Insight Into the Differences Between Slow-Twitch Muscle and Fast-Twitch Muscle of Pseudocaranx dentex Open
Fish skeletal muscles are mainly composed of two distinct types, fast-twitch and slow-twitch muscles, and they play important roles in maintaining movement and energy metabolism. The fast-twitch muscle contracts quickly and is mainly respo…
View article: Effective factors of peroral endoscopic myotomy for treatment of achalasia.
Effective factors of peroral endoscopic myotomy for treatment of achalasia. Open
POEM is safe, feasible and effective for AC treatment, based on factors such as drinking history, Eckardt score 1 year after the operation, Hb, ALT, WBC, IL-6, and hs-CRP.
View article: Localization with Sampling-Argmax
Localization with Sampling-Argmax Open
Soft-argmax operation is commonly adopted in detection-based methods to localize the target position in a differentiable manner. However, training the neural network with soft-argmax makes the shape of the probability map unconstrained. Co…
View article: Human Pose Regression with Residual Log-likelihood Estimation
Human Pose Regression with Residual Log-likelihood Estimation Open
Heatmap-based methods dominate in the field of human pose estimation by modelling the output distribution through likelihood heatmaps. In contrast, regression-based methods are more efficient but suffer from inferior performance. In this w…
View article: ArtiBoost: Boosting Articulated 3D Hand-Object Pose Estimation via Online Exploration and Synthesis
ArtiBoost: Boosting Articulated 3D Hand-Object Pose Estimation via Online Exploration and Synthesis Open
Estimating the articulated 3D hand-object pose from a single RGB image is a highly ambiguous and challenging problem, requiring large-scale datasets that contain diverse hand poses, object types, and camera viewpoints. Most real-world data…
View article: TDAF: Top-Down Attention Framework for Vision Tasks
TDAF: Top-Down Attention Framework for Vision Tasks Open
Human attention mechanisms often work in a top-down manner, yet it is not well explored in vision research. Here, we propose the Top-Down Attention Framework (TDAF) to capture top-down attentions, which can be easily adopted in most existi…
View article: Segmented Cantilever and Array Configurations for Wider Frequency Band and Higher Power Generation in Piezoelectric Vibration Energy Harvester
Segmented Cantilever and Array Configurations for Wider Frequency Band and Higher Power Generation in Piezoelectric Vibration Energy Harvester Open
This letter reports a piezoelectric vibration energy harvester which energy conversion efficiency is significantly improved by arraying piezoelectric sheets on cantilever beams, and the operation frequency band is widened by applying two‐s…
View article: TDAF: Top-Down Attention Framework for Vision Tasks
TDAF: Top-Down Attention Framework for Vision Tasks Open
Human attention mechanisms often work in a top-down manner, yet it is not well explored in vision research. Here, we propose the Top-Down Attention Framework (TDAF) to capture top-down attentions, which can be easily adopted in most existi…
View article: CPF: Learning a Contact Potential Field to Model the Hand-Object Interaction
CPF: Learning a Contact Potential Field to Model the Hand-Object Interaction Open
Modeling the hand-object (HO) interaction not only requires estimation of the HO pose, but also pays attention to the contact due to their interaction. Significant progress has been made in estimating hand and object separately with deep l…
View article: HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation
HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation Open
Model-based 3D pose and shape estimation methods reconstruct a full 3D mesh for the human body by estimating several parameters. However, learning the abstract parameters is a highly non-linear process and suffers from image-model misalign…