Visual-Intelligence
@VI_Journal_CSIG
Followers
88
Following
685
Media
10
Statuses
367
Official journal of China Society of Image and Graphics (CSIG). The jouarnl is published by Springer, sponsored by CSIG. E-ISSN 2731-9008.
Beijing
Joined February 2025
🚀Check out the new paper! 🚀Authors from University of Montreal, Woods Hole Oceanographic Institution, and UC Berkeley proposed DASFormer, a novel self-supervised pretraining technique on DAS data for next-generation earthquake monitoring. @SpringerEng.
0
0
3
🔥Check out DreamConnect! By integrating an asynchronous diffusion strategy, this framework establishes an effective interface with human “dreams”, and progressively refines their final image synthesis. #Braintoimagegeneration.@SpringerEng.
0
1
1
🚀Check out the latest research🚀Authors from BIT proposed DehazeMamba, a large multi-modal model (LMM) guided, Mamba-based image dehazing method. @SpringerEng @SpringerCompSci .#Imagedehazing #LMM #Mamba.
0
0
2
🚀Checkout our latest paper! 🚀Authors from ETH Zürich investigated the application and performance of the Segment Anything Model 2 (SAM2) in the challenging task of video camouflaged object segmentation. @SpringerEng @SpringerCompSci .#MLLM #SAM2.
0
0
5
🎉🎉Good News: Visual Intelligence has been accepted to be indexed in Ei Compendex. We'd like to take this opportunity to thank all authors, reviewers, readers and editorial board members for the continuous support to the journal. @SpringerEng @SciOpenTUP .#Ei #journalindexing.
0
1
4
🔥Existing LMMs face the following problems: Corpus limitations; 2) Inconsistent understanding and generation capabilities across modalities; 3) Lack of true multi-modality. 🔗#AI #LargeAImodels #LLM #GPT.
0
0
2
Congratulations to Dr. Jingyu Yu, our journal's Associate Editor-in-Chief, and his team for winning the Best Paper Award at SIGGRAPH 2025! CAST: Component-Aligned 3D Scene Reconstruction From an RGB Image. See the news at: .
0
0
1
🔥MLLMs' ability to detect subtle visual spoofing and forgery clues in face attack detection tasks remains underexplored. Check out this benchmark, SHIELD, to evaluate MLLMs for face spoofing and forgery detection. @SpringerEng @SpringerCompSci.
0
2
2
🔥Check out this cover paper of Visual Intelligence journal! It proposes FusionMamba, aiming at overcoming the challenges faced by CNNs and Vision Transformers (ViTs) in computer vision tasks. #Multimodal #Imagefusion #Mamba .@SpringerEng @SpringerCompSci
0
0
3
🔥Check out the new method! The unique advantage of the divide-and-conquer reconstruction method is its ability to provide tailored defense models against different types of adversarial attacks. #Adversarialexampledefense.
0
1
2
🔥New research: Check out this high-resolution (1280 x 1280) X-ray image based pre-trained baseline model! The authors collected a large-scale dataset containing more than 1 million X-ray images. #Medicalreportgeneration.@SpringerEng @SpringerCompSci.
0
0
1
✨Existing panoptic segmentation methods assume the differences between instances are greater than those between points belonging to the same instance and use heuristic techniques for segmentation. Check out this paper's new idea. #Pointcloud #Lidar.
0
0
2
🥳🥳🥳Congratulations to our esteemed Associate Editors for being elevated to IEEE Fellow Class of 2025! Their outstanding contributions continue to inspire and elevate the academic community. #IEEEFellow #AcademicExcellence
0
0
2
🌟Visual Intelligence highly cited papers No. 10🌟.This paper tackles the semantic change detection and land cover mapping tasks simultaneously by proposing a coarse-to-fine attention tree (CAT) model. #Finegrainedrecognition #Coarsetofine #attentiontree.
0
0
2
🌟Visual Intelligence highly cited papers No. 9🌟.In addition to the unavoidable noise, the uneven spatial distribution of light and dark can further make the problem of vehicle detection in dim light more difficult. #Multiexposureandfusion.
0
0
4
🌟Visual Intelligence highly cited papers No. 8🌟.The authors proposed a simple approach that combines recurrent neural networks (RNNs) and attention mechanism for motion prediction. Read more: #Humanmotionprediction #RNN.
0
0
1
🌟Visual Intelligence highly cited papers No. 7🌟.Check out this cross-camera self-distillation framework for unsupervised person Re-ID to alleviate the effect of camera variation. 🔗Read more: #Unsupervisedlearning #Personreid #Knowledgedistillation.
0
0
2
🌟Visual Intelligence highly cited papers No. 6🌟.Deep learning models are characterized by catastrophic forgetting of old knowledge when learning new classes. This study proposed a Bayesian generative model to address this issue. 🔗#Continuallearning #AI.
0
0
1
🌟Visual Intelligence highly cited papers No. 5🌟.Intelligent video coding includes visual content compact representation models and methods at different granularity levels and in different areas. #AI #generativecoding.
0
0
2
🌟Visual Intelligence highly cited papers No. 4🌟.Check out this novel semantic-aware knowledge distillation scheme to fill in the dimension gap between intermediate features of teacher and student. 🔗Read more: #Knowledgedistillation #Selfattention #llm.
0
2
5