15 citations · 46 across the 13 of their papers we have counts for
14 papers
DNAEdit: Direct Noise Alignment for Text-Guided Rectified Flow Editing
Chenxi Xie, Minghan Li, Shuai Li +3
Leveraging the powerful generation capability of large-scale pretrained text-to-image models, training-free methods have demonstrated impressive image editing results. Conventional…
UniVS: Unified and Universal Video Segmentation with Prompts as Queries
Minghan Li, Shuai Li, Xindong Zhang +1
Despite the recent advances in unified image segmentation (IS), developing a unified video segmentation (VS) model remains a challenge. This is mainly because generic category-spec…
OpenSD: Unified Open-Vocabulary Segmentation and Detection
Shuai Li, Minghan Li, Pengfei Wang +1
Recently, a few open-vocabulary methods have been proposed by employing a unified architecture to tackle generic segmentation and detection tasks. However, their performance still…
MDQE: Mining Discriminative Query Embeddings to Segment Occluded Instances on Challenging Videos
Minghan Li, Shuai Li, Wangmeng Xiang +1
While impressive progress has been achieved, video instance segmentation (VIS) methods with per-clip input often fail on challenging videos with occluded objects and crowded scenes…
One-to-Few Label Assignment for End-to-End Dense Detection
Shuai Li, Minghan Li, Ruihuang Li +2
One-to-one (o2o) label assignment plays a key role for transformer based end-to-end detection, and it has been recently introduced in fully convolutional detectors for end-to-end d…
MSF: Motion-guided Sequential Fusion for Efficient 3D Object Detection from Point Cloud Sequences
Chenhang He, Ruihuang Li, Yabin Zhang +2
Point cloud sequences are commonly used to accurately detect 3D objects in applications such as autonomous driving. Current top-performing multi-frame detectors mostly follow a Det…
SIM: Semantic-aware Instance Mask Generation for Box-Supervised Instance Segmentation
Ruihuang Li, Chenhang He, Yabin Zhang +3
Weakly supervised instance segmentation using only bounding box annotations has recently attracted much research attention. Most of the current efforts leverage low-level image fea…
DynaMask: Dynamic Mask Selection for Instance Segmentation
Ruihuang Li, Chenhang He, Shuai Li +2
The representative instance segmentation methods mostly segment different object instances with a mask of the fixed resolution, e.g., 28*28 grid. However, a low-resolution mask los…
Voxel Set Transformer: A Set-to-Set Approach to 3D Object Detection from Point Clouds
Chenhang He, Ruihuang Li, Shuai Li +1
Transformer has demonstrated promising performance in many 2D vision tasks. However, it is cumbersome to compute the self-attention on large-scale point cloud data because point cl…
Towards Robust 2D Convolution for Reliable Visual Recognition
Lida Li, Shuai Li, Kun Wang +2
2D convolution (Conv2d), which is responsible for extracting features from the input image, is one of the key modules of a convolutional neural network (CNN). However, Conv2d is vu…
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
Ruihuang Li, Shuai Li, Chenhang He +3
Domain adaptive semantic segmentation aims to learn a model with the supervision of source domain data, and produce satisfactory dense predictions on unlabeled target domain. One p…
A Dual Weighting Label Assignment Scheme for Object Detection
Shuai Li, Chenhang He, Ruihuang Li +1
Label assignment (LA), which aims to assign each training sample a positive (pos) and a negative (neg) loss weight, plays an important role in object detection. Existing LA methods…
Semi-supervised Models are Strong Unsupervised Domain Adaptation Learners
Yabin Zhang, Haojian Zhang, Bin Deng +3
Unsupervised domain adaptation (UDA) and semi-supervised learning (SSL) are two typical strategies to reduce expensive manual annotations in machine learning. In order to learn eff…
Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation
Minghan Li, Shuai Li, Lida Li +1
Modern one-stage video instance segmentation networks suffer from two limitations. First, convolutional features are neither aligned with anchor boxes nor with ground-truth boundin…