Search Results - "Niu, Yulei"

Unbiased Scene Graph Generation From Biased Training by Tang, Kaihua, Niu, Yulei, Huang, Jianqiang, Shi, Jiaxin, Zhang, Hanwang

“…Today's scene graph generation (SGG) task is still far from practical, mainly due to the severe training bias, e.g., collapsing diverse "human walk on / sit on…”
Get full text

Multi-Modal Multi-Scale Deep Learning for Large-Scale Image Annotation by Niu, Yulei, Lu, Zhiwu, Wen, Ji-Rong, Xiang, Tao, Chang, Shih-Fu

“…Image annotation aims to annotate a given image with a variable number of class labels corresponding to diverse visual concepts. In this paper, we address two…”
Get full text

A Study on the Effect of Initial Temperature on Combustion Characteristics of RDX Based on the Optical Diagnosis Methods by Yan, Zhiyu, Song, Chen, Li, Qianqian, Niu, Yulei, Wang, Jinhua, Huang, Zuohua

“…The purpose of this work is to investigate the effect of different initial temperatures on the combustion characteristics of RDX materials. In the experiments,…”
Get full text

Grounding Referring Expressions in Images by Variational Context by Zhang, Hanwang, Niu, Yulei, Chang, Shih-Fu

“…We focus on grounding (i.e., localizing or linking) referring expressions in images, e.g., "largest elephant standing behind baby elephant". This is a general…”
Get full text

Two Causal Principles for Improving Visual Dialog by Qi, Jiaxin, Niu, Yulei, Huang, Jianqiang, Zhang, Hanwang

“…This paper unravels the design tricks adopted by us, the champion team MReaL-BDAI, for Visual Dialog Challenge 2019: two causal principles for improving Visual…”
Get full text

Counterfactual VQA: A Cause-Effect Look at Language Bias by Niu, Yulei, Tang, Kaihua, Zhang, Hanwang, Lu, Zhiwu, Hua, Xian-Sheng, Wen, Ji-Rong

“…VQA models may tend to rely on language bias as a shortcut and thus fail to sufficiently learn the multi-modal knowledge from both vision and language. Recent…”
Get full text

Counterfactual Samples Synthesizing and Training for Robust Visual Question Answering by Chen, Long, Zheng, Yuhang, Niu, Yulei, Zhang, Hanwang, Xiao, Jun

“…Today's VQA models still tend to capture superficial linguistic correlations in the training set and fail to generalize to the test set with different QA…”
Get full text

Recursive Visual Attention in Visual Dialog by Niu, Yulei, Zhang, Hanwang, Zhang, Manli, Zhang, Jianhong, Lu, Zhiwu, Wen, Ji-Rong

“…Visual dialog is a challenging vision-language task, which requires the agent to answer multi-round questions about an image. It typically needs to address two…”
Get full text

Variational Context: Exploiting Visual and Textual Context for Grounding Referring Expressions by Niu, Yulei, Zhang, Hanwang, Lu, Zhiwu, Chang, Shih-Fu

“…We focus on grounding (i.e., localizing or linking) referring expressions in images, e.g., "largest elephant standing behind baby elephant". This is a general…”
Get full text

Domain-Adaptive Few-Shot Learning by Zhao, An, Ding, Mingyu, Lu, Zhiwu, Xiang, Tao, Niu, Yulei, Guan, Jiechao, Wen, Ji-Rong

“…Existing few-shot learning (FSL) methods make the implicit assumption that the few target class samples are from the same domain as the source class samples…”
Get full text

Energy output performance of aluminized explosive containing Al/PTFE reactive materials by Jiang, Fan, Sun, Peipei, Bu, Yufan, Niu, Yulei, Li, Yuanyuan, Zhang, Kun, Wang, Xiaofeng, Nan, Hai

“…In this paper, a series of CL-20 based explosive formulations containing Al/PTFE reactive materials are designed using a self-designed closed explosion test…”
Get full text

Classification-Then-Grounding: Reformulating Video Scene Graphs as Temporal Bipartite Graphs by Gao, Kaifeng, Chen, Long, Niu, Yulei, Shao, Jian, Xiao, Jun

“…Today's VidSGG models are all proposal-based methods, i.e., they first generate numerous paired subject-object snippets as proposals, and then conduct…”
Get full text

Decoupling heat-pressure potential energy of compressed air energy storage system: Using near-isothermal compressing and thermal energy storage by Li, Ruixiong, Tao, Rui, Yao, Erren, Zhang, Haoran, Niu, Yulei, Ling, Lanning, Yan, An, Wang, Huanran

“…Compressed air energy storage (CAES) system is a promising solution for matching the intermittent renewable energy sources and stable electricity demand of end…”
Get full text

DiGeo: Discriminative Geometry-Aware Learning for Generalized Few-Shot Object Detection by Ma, Jiawei, Niu, Yulei, Xu, Jincheng, Huang, Shiyuan, Han, Guangxing, Chang, Shih-Fu

“…Generalized few-shot object detection aims to achieve precise detection on both base classes with abundant annotations and novel classes with limited training…”
Get full text

Preparation of Ultrafine CL‐20 by Wet Grinding Method and the Study on Its Properties by Li, Yuanyuan, Ni, Bing, Li, Kun, Nan, Hai, Niu, Yulei, Jiang, Fan, Li, Xin

“…In this paper, ultrafine CL‐20 was prepared by wet grinding procedure. The morphology, particle size, sample composition, and crystal structure of CL‐20 before…”
Get full text

Multiobjective Optimization of a Combined Heating and Power System Based on Compressed-Air Energy Storage and Thermochemical Technology by Yao, Erren, Zhong, Like, Li, Ruixiong, Niu, Yulei, Wang, Huanran, Xi, Guang

“…AbstractCompressed-air energy storage has been considered as a promising technology to smooth the fluctuations of renewable energy sources and improve the…”
Get full text

Graph-boosted convolutional neural networks for semantic segmentation by Guangzhen Liu, Peng Han, Yulei Niu, Wenwu Yuan, Zhiwu Lu, Ji-Rong Wen

“…This paper investigates the problem of weakly-supervised semantic segmentation, where image-level labels are used as weak supervision. Inspired by the…”
Get full text

Introspective Distillation for Robust Question Answering by Niu, Yulei, Zhang, Hanwang

“…Question answering (QA) models are well-known to exploit data bias, e.g., the language prior in visual QA and the position bias in reading comprehension…”
Get full text

RAP: Retrieval-Augmented Planner for Adaptive Procedure Planning in Instructional Videos by Zare, Ali, Niu, Yulei, Ayyubi, Hammad, Chang, Shih-fu

“…Procedure Planning in instructional videos entails generating a sequence of action steps based on visual observations of the initial and target states. Despite…”
Get full text

Debiased Fine-Tuning for Vision-language Models by Prompt Regularization by Zhu, Beier, Niu, Yulei, Lee, Saeil, Hur, Minhoe, Zhang, Hanwang

“…We present a new paradigm for fine-tuning large-scale visionlanguage pre-trained models on downstream task, dubbed Prompt Regularization (ProReg). Different…”
Get full text