User profiles for Weifeng Ge
 | Weifeng GeFudan University Verified email at fudan.edu.cn Cited by 1856 |
Deep metric learning with hierarchical triplet loss
W Ge - Proceedings of the European conference on …, 2018 - openaccess.thecvf.com
We present a novel hierarchical triplet loss (HTL) capable of automatically collecting
informative training samples (triplets) via a defined hierarchical tree that encodes global context …
informative training samples (triplets) via a defined hierarchical tree that encodes global context …
Weakly supervised complementary parts models for fine-grained image classification from the bottom up
Given a training dataset composed of images and corresponding category labels, deep
convolutional neural networks show a strong ability in mining discriminative parts for image …
convolutional neural networks show a strong ability in mining discriminative parts for image …
A systematic review on affective computing: Emotion models, databases, and recent advances
Affective computing conjoins the research topics of emotion recognition and sentiment
analysis, and can be realized with unimodal or multimodal data, consisting primarily of physical …
analysis, and can be realized with unimodal or multimodal data, consisting primarily of physical …
Borrowing treasures from the wealthy: Deep transfer learning through selective joint fine-tuning
Deep neural networks require a large amount of labeled training data during supervised
learning. However, collecting and labeling so much data might be infeasible in many cases. In …
learning. However, collecting and labeling so much data might be infeasible in many cases. In …
Multi-evidence filtering and fusion for multi-label classification, object detection and semantic segmentation based on weakly supervised learning
Supervised object detection and semantic segmentation require object or even pixel level
annotations. When there exist image level labels only, it is challenging for weakly supervised …
annotations. When there exist image level labels only, it is challenging for weakly supervised …
Ferv39k: A large-scale multi-scene dataset for facial expression recognition in videos
Current benchmarks for facial expression recognition (FER) mainly focus on static images,
while there are limited datasets for FER in videos. It is still ambiguous to evaluate whether …
while there are limited datasets for FER in videos. It is still ambiguous to evaluate whether …
Digital twin-driven fault diagnosis method for composite faults by combining virtual and real data
The subsea production system is essential for the subsea production of oil and gas. Real-time
monitoring can ensure safe production. The subsea production control system is the core …
monitoring can ensure safe production. The subsea production control system is the core …
GraphFPN: Graph feature pyramid network for object detection
Feature pyramids have been proven powerful in image understanding tasks that require
multi-scale features. Stateof-the-art methods for multi-scale feature learning focus on performing …
multi-scale features. Stateof-the-art methods for multi-scale feature learning focus on performing …
Remaining useful life estimation of structure systems under the influence of multiple causes: Subsea pipelines as a case study
…, Y Liu, X Kong, H Wang, H Xu, W Ge - IEEE Transactions on …, 2019 - ieeexplore.ieee.org
In dynamic complex environments, the degradation of structure systems is generally caused
not by a single factor but by multiple ones, and the process is subject to a high level of …
not by a single factor but by multiple ones, and the process is subject to a high level of …
Attribute surrogates learning and spectral tokens pooling in transformers for few-shot learning
This paper presents new hierarchically cascaded transformers that can improve data efficiency
through attribute surrogates learning and spectral tokens pooling. Vision transformers …
through attribute surrogates learning and spectral tokens pooling. Vision transformers …