Wenqi Zhu

Wenqi Zhu

Auto-generated profile, based on publicly available information

Wenqi Zhu

School of Electrical and Information Engineering, Tianjin University, Tianjin, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
7 October 2024 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
WZWenqi Zhu
Wenqi Zhu
JCJiale Cao
Jiale Cao
JXJin Xie
Jin Xie
Shuangming YangShuangming Yang
Shuangming Yang
YPYanwei Pang
Yanwei Pang
Open-vocabulary video instance segmentation strives to segment and track instances belonging to an open set of categories in a videos. The vision-language model Contrastive Language-Image Pre-training (CLIP) has shown robust zero-shot classification ability in image-level open-vocabulary tasks. In this paper, we propose a simple encoder-decoder network, called CLIP-VIS, to adapt CLIP for open-vocabulary video instance segmentation. Our CLIP-VIS adopts frozen CLIP and introduces three modules, including class-agnostic mask generation, temporal top K -enhanced matching, and weighted open-vocabulary classification. Given a set of initial queries, class-agnostic mask generation introduces a pixel decoder and a transformer decoder on CLIP pre-trained image encoder to predict query masks and corresponding object scores and mask IoU scores. Then, temporal top K -enhanced matching performs query matching across frames using the K mostly matched frames. Finally, weighted open-vocabulary classification first employs mask pooling to generate query visual features from CLIP pre-trained image encoder, and second performs weighted classification using object scores and mask IoU scores. Our CLIP-VIS does not require the annotations of instance categories and identities. The experiments are performed on various video instance segmentation datasets, which demonstrate the effectiveness of our proposed method, especially for novel categories. When using ConvNeXt-B as backbone, our CLIP-VIS achieves the AP and APn scores of 32.2% and 40.2% on the validation set of LV-VIS dataset, which outperforms OV2Seg by 11.1% and 23.9% respectively. We will release the source code and models at https://github.com/zwq456/CLIP-VIS.git.

1 December 2023 in Journal of Materials Processing Technology
Verified
Journal Article
Closed access
BWBaolei Wu
Baolei Wu
WYWeiyuan Yu
Weiyuan Yu
WZWenqi Zhu
Wenqi Zhu
YLYang Li
Yang Li
Zhen-Zhong HuZhen-Zhong Hu
Zhen-Zhong Hu

31 October 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
WZWenqi Zhu
Wenqi Zhu
BXBin Xie
Bin Xie
JCJiale Cao
Jiale Cao
JXJin Xie
Jin Xie
FKFahad Shahbaz Khan
Fahad Shahbaz Khan
YPYanwei Pang
Yanwei Pang
Open-vocabulary semantic segmentation aims to partition an image into distinct semantic regions based on an open set of categories. Existing approaches primarily rely on image-level pre-trained vision-language models to perform this pixel-level task. In this paper, we propose SED, a simple yet effective encoder-decoder architecture for open-vocabulary semantic segmentation leveraging pre-trained vision-language models. SED consists of a hierarchical image encoder, a text encoder, and a gradual fusion decoder. The hierarchical image encoder and text encoder collaboratively generate a cost volume, which is progressively decoded by the gradual fusion decoder to produce segmentation results. In contrast to a plain encoder, the hierarchical encoder better captures image detail information while maintaining linear computational complexity with respect to input size. The gradual fusion decoder adopts a top-down structure to progressively integrate high-resolution features with the cost volume. Furthermore, a category early rejection strategy is introduced in gradual fusion decoder to filter out non-existent categories at different layers, significantly improving inference efficiency. Based on SED, we further introduce two modules, including non-label text embedding and additional category early rejection in the encoder. Moreover, we extend our method with minimal decoder modification for open-vocabulary video semantic segmentation. Extensive experiments on multiple datasets validate the effectiveness and efficiency of our proposed method. With ConvNeXt-B, our method achieves an mIoU of 34.9% on the ADE20 K with 150 classes (i.e., A-150) at an inference speed of 69 ms per image on a single A6000 GPU, and has an mIoU score of 40.2% on video segmentation dataset VSPW.

21 April 2023
Verified
Conference Paper
Closed access
WZWenqi Zhu
Wenqi Zhu
JCJunzhe Cao
Junzhe Cao
Genome-wide prediction methods establish a genetic statistical model using molecular marker genotypes distributed throughout the whole genome and biological trait data to predict phenotype values, in order to providing a guide for individual selection breeding. Most of the existing genome-wide selection methods focus on the prediction of a single trait, which can only examine each trait one by one when predicting multiple traits that would translate into neglect of the correlation among traits. In this paper, we proposed a joint prediction method for multiple biological traits based on feedback stacking deep networks, which stack orderly the output of the model to the non-shared layer so as to introduce the correlation among traits into the learning network. The method selected the single nucleotide polymorphism (SNP) loci obtained from whole genome sequencing as molecular markers, extracted the genotype of each SNP allele to constitute sample features, and adopted a two-stage feature selection method to pick up the SNP loci highly correlated with target, and then constructed a deep network model with feedback stacking for joint prediction of multiple traits. The proposed approach was compared with other models by using oyster and corn datasets, and the results showed that the method on this work could achieve better prediction effect than the others.