Qi Tian
YOU?
Author Swipe
View article: PSR: Scaling Multi-Subject Personalized Image Generation with Pairwise Subject-Consistency Rewards
PSR: Scaling Multi-Subject Personalized Image Generation with Pairwise Subject-Consistency Rewards Open
Personalized generation models for a single subject have demonstrated remarkable effectiveness, highlighting their significant potential. However, when extended to multiple subjects, existing models often exhibit degraded performance, part…
View article: Observations on the Efficacy and Safety of FRED™ Jr FD for the Treatment of Intracranial Unruptured Distal Aneurysms: A Single-Center Experience
Observations on the Efficacy and Safety of FRED™ Jr FD for the Treatment of Intracranial Unruptured Distal Aneurysms: A Single-Center Experience Open
OKM＝O'Kelly-Marotta grading scale.
View article: Fog computing based cost optimization for university governance
Fog computing based cost optimization for university governance Open
This study presented a new architecture based on fog computing to effectively reduce the burdensome cost of university governance. The process is established by enhancing network performance and optimizing resource utilization. The solutio…
View article: Snap-Snap: Taking Two Images to Reconstruct 3D Human Gaussians in Milliseconds
Snap-Snap: Taking Two Images to Reconstruct 3D Human Gaussians in Milliseconds Open
Reconstructing 3D human bodies from sparse views has been an appealing topic, which is crucial to broader the related applications. In this paper, we propose a quite challenging but valuable task to reconstruct the human body from only two…
View article: Bone marrow adipocytes: key players in vascular niches, aging, and disease
Bone marrow adipocytes: key players in vascular niches, aging, and disease Open
Bone marrow adipocytes (BMAs) are emerging as metabolically active endocrine organs within the bone marrow microenvironment, engaging in extensive crosstalk with vascular niches, osteogenic cells, and hematopoietic compartments. In aging a…
View article: Efficient Multi-modal Long Context Learning for Training-free Adaptation
Efficient Multi-modal Long Context Learning for Training-free Adaptation Open
Traditional approaches to adapting multi-modal large language models (MLLMs) to new tasks have relied heavily on fine-tuning. This paper introduces Efficient Multi-Modal Long Context Learning (EMLoC), a novel training-free alternative that…
View article: Infinite-Canvas: Higher-Resolution Video Outpainting with Extensive Content Generation
Infinite-Canvas: Higher-Resolution Video Outpainting with Extensive Content Generation Open
This paper explores higher-resolution video outpainting with extensive content generation. We point out common issues faced by existing methods when attempting to largely outpaint videos: the generation of low-quality content and limitatio…
View article: GraphATC: advancing multilevel and multi-label anatomical therapeutic chemical classification via atom-level graph learning
GraphATC: advancing multilevel and multi-label anatomical therapeutic chemical classification via atom-level graph learning Open
The accurate categorization of compounds within the anatomical therapeutic chemical (ATC) system is fundamental for drug development and fundamental research. Although this area has garnered significant research focus for over a decade, th…
View article: Biomechanical insights and optimization in the teaching design of badminton games based on motion capture and adaptive virtual reality video coding
Biomechanical insights and optimization in the teaching design of badminton games based on motion capture and adaptive virtual reality video coding Open
The application of games in sports not only brings new development to sports, but also brings new requirements to sports. To maintain and enhance students’ learning motivation and interest, more effective individualized teaching is needed.…
View article: CDG: A semantic–structural controllable framework for crack image generation in complex scenes
CDG: A semantic–structural controllable framework for crack image generation in complex scenes Open
View article: LiftImage3D: Lifting Any Single Image to 3D Gaussians with Video Generation Priors
LiftImage3D: Lifting Any Single Image to 3D Gaussians with Video Generation Priors Open
Single-image 3D reconstruction remains a fundamental challenge in computer vision due to inherent geometric ambiguities and limited viewpoint information. Recent advances in Latent Video Diffusion Models (LVDMs) offer promising 3D priors l…
View article: Advances in Plant GABA Research: Biological Functions, Synthesis Mechanisms and Regulatory Pathways
Advances in Plant GABA Research: Biological Functions, Synthesis Mechanisms and Regulatory Pathways Open
The γ-aminobutyric acid (GABA) is a widely distributed neurotransmitter in living organisms, known for its inhibitory role in animals. GABA exerts calming effects on the mind, lowers blood pressure in animals, and enhances stress resistanc…
View article: Frailty increases depression risk independently of cognitive decline: Insights from Mendelian randomization and cross-sectional analysis
Frailty increases depression risk independently of cognitive decline: Insights from Mendelian randomization and cross-sectional analysis Open
We provide evidence that frailty could increase depression risk independently of cognitive decline. Further research with a larger sample size is necessary.
View article: Follow-Your-Canvas: Higher-Resolution Video Outpainting with Extensive Content Generation
Follow-Your-Canvas: Higher-Resolution Video Outpainting with Extensive Content Generation Open
This paper explores higher-resolution video outpainting with extensive content generation. We point out common issues faced by existing methods when attempting to largely outpaint videos: the generation of low-quality content and limitatio…
View article: Predicting the sea lamprey population dynamic and ecosystem stability based on sex ratio corrected logistic model
Predicting the sea lamprey population dynamic and ecosystem stability based on sex ratio corrected logistic model Open
The sex ratio of the sea lamprey(Petromyzon marinus) is influenced by population density. A sex ratio corrected population model and a ecosystem stability model were established based on the logistic population model. In these models, the …
View article: Segment Any 4D Gaussians
Segment Any 4D Gaussians Open
Modeling, understanding, and reconstructing the real world are crucial in XR/VR. Recently, 3D Gaussian Splatting (3D-GS) methods have shown remarkable success in modeling and understanding 3D scenes. Similarly, various 4D representations h…
View article: SDPT: Semantic-Aware Dimension-Pooling Transformer for Image Segmentation
SDPT: Semantic-Aware Dimension-Pooling Transformer for Image Segmentation Open
Image segmentation plays a critical role in autonomous driving by providing vehicles with a detailed and accurate understanding of their surroundings. Transformers have recently shown encouraging results in image segmentation. However, tra…
View article: A Survey of Generative Techniques for Spatial-Temporal Data Mining
A Survey of Generative Techniques for Spatial-Temporal Data Mining Open
This paper focuses on the integration of generative techniques into spatial-temporal data mining, considering the significant growth and diverse nature of spatial-temporal data. With the advancements in RNNs, CNNs, and other non-generative…
View article: Predicting dust pollution from dry bulk ports in coastal cities: A hybrid approach based on data decomposition and deep learning
Predicting dust pollution from dry bulk ports in coastal cities: A hybrid approach based on data decomposition and deep learning Open
Dust pollution from storage and handling of materials in dry bulk ports seriously affects air quality and public health in coastal cities. Accurate prediction of dust pollution helps identify risks early and take preventive measures. Howev…
View article: Visual Tuning
Visual Tuning Open
Fine-tuning visual models has been widely shown promising performance on many downstream visual tasks. With the surprising development of pre-trained visual foundation models, visual tuning jumped out of the standard modus operandi that fi…
View article: AlignZeg: Mitigating Objective Misalignment for Zero-shot Semantic Segmentation
AlignZeg: Mitigating Objective Misalignment for Zero-shot Semantic Segmentation Open
A serious issue that harms the performance of zero-shot visual recognition is named objective misalignment, i.e., the learning objective prioritizes improving the recognition accuracy of seen classes rather than unseen classes, while the l…
View article: BLADE: Enhancing Black-box Large Language Models with Small Domain-Specific Models
BLADE: Enhancing Black-box Large Language Models with Small Domain-Specific Models Open
Large Language Models (LLMs) like ChatGPT and GPT-4 are versatile and capable of addressing a diverse range of tasks. However, general LLMs, which are developed on open-domain data, may lack the domain-specific knowledge essential for task…
View article: LION: Implicit Vision Prompt Tuning
LION: Implicit Vision Prompt Tuning Open
Despite recent promising performances across a range of vision tasks, vision Transformers still have an issue of high computational costs. Recently, vision prompt learning has provided an economical solution to this problem without fine-tu…
View article: Influence of distributed generation on fault characteristics and relay protection of rural distribution network
Influence of distributed generation on fault characteristics and relay protection of rural distribution network Open
High proportion distributed generation affect traditional rural distribution network. In the paper, a typical line model of rural distribution network with high proportion distributed generation was constructed, and the current and voltage…
View article: Deep Watermarking for Deep Intellectual Property Protection: A Comprehensive Survey
Deep Watermarking for Deep Intellectual Property Protection: A Comprehensive Survey Open
View article: When Parameter-efficient Tuning Meets General-purpose Vision-language Models
When Parameter-efficient Tuning Meets General-purpose Vision-language Models Open
Instruction tuning has shown promising potential for developing general-purpose AI capabilities by using large-scale pre-trained models and boosts growing research to integrate multimodal information for creative applications. However, exi…
View article: Parameter Efficient Fine-tuning via Cross Block Orchestration for Segment Anything Model
Parameter Efficient Fine-tuning via Cross Block Orchestration for Segment Anything Model Open
Parameter-efficient fine-tuning (PEFT) is an effective methodology to unleash the potential of large foundation models in novel scenarios with limited training data. In the computer vision community, PEFT has shown effectiveness in image c…
View article: One-Bit Supervision for Image Classification: Problem, Solution, and Beyond
One-Bit Supervision for Image Classification: Problem, Solution, and Beyond Open
This article presents one-bit supervision, a novel setting of learning with fewer labels, for image classification. Instead of the training model using the accurate label of each sample, our setting requires the model to interact with the …
View article: Probabilistic Tree-of-thought Reasoning for Answering Knowledge-intensive Complex Questions
Probabilistic Tree-of-thought Reasoning for Answering Knowledge-intensive Complex Questions Open
Large language models (LLMs) are capable of answering knowledge-intensive complex questions with chain-of-thought (CoT) reasoning. However, they tend to generate factually incorrect reasoning steps when the required knowledge is not availa…
View article: AiluRus: A Scalable ViT Framework for Dense Prediction
AiluRus: A Scalable ViT Framework for Dense Prediction Open
Vision transformers (ViTs) have emerged as a prevalent architecture for vision tasks owing to their impressive performance. However, when it comes to handling long token sequences, especially in dense prediction tasks that require high-res…