activity
20202023
most citedMask Matching Transformer for Few-Shot Segmentation

20 citations · 68 across the 8 of their papers we have counts for

collaborators
Showing cs.CVShow all

8 papers · 1 filter

cs.CV2023★ 5 cited

SLCA: Slow Learner with Classifier Alignment for Continual Learning on a Pre-trained Model

Gengwei Zhang, Liyuan Wang, Guoliang Kang +2

The goal of continual learning is to improve the performance of recognition models in learning sequentially arrived data. Although most existing works are established on the premis…

cs.CV2022★ 20 cited

Mask Matching Transformer for Few-Shot Segmentation

Siyu Jiao, Gengwei Zhang, Shant Navasardyan +4

In this paper, we aim to tackle the challenging few-shot segmentation task from a new perspective. Typical methods follow the paradigm to firstly learn prototypical features from s…

cs.CV2022★ 2 cited

Continual Object Detection via Prototypical Task Correlation Guided Gating Mechanism

Binbin Yang, Xinchi Deng, Han Shi +6

Continual learning is a challenging real-world problem for constructing a mature AI system when data are provided in a streaming fashion. Despite recent progress in continual class…

cs.CV2021

NASOA: Towards Faster Task-oriented Online Fine-tuning with a Zoo of Models

Hang Xu, Ning Kang, Gengwei Zhang +3

Fine-tuning from pre-trained ImageNet models has been a simple, effective, and popular approach for various computer vision tasks. The common practice of fine-tuning is to adopt a…

cs.CV2021★ 16 cited

Loss Function Discovery for Object Detection via Convergence-Simulation Driven Search

Peidong Liu, Gengwei Zhang, Bochao Wang +4

Designing proper loss functions for vision tasks has been a long-standing research direction to advance the capability of existing models. For object detection, the well-establishe…

cs.CV2020

Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation

Gengwei Zhang, Yiming Gao, Hang Xu +3

Panoptic segmentation that unifies instance segmentation and semantic segmentation has recently attracted increasing attention. While most existing methods focus on designing novel…