iSeg improves image segmentation accuracy with less training data.
Lin Sun, Jiale Cao, Jin Xie
― 4 min read
Cutting edge science explained simply
iSeg improves image segmentation accuracy with less training data.
Lin Sun, Jiale Cao, Jin Xie
― 4 min read
A new model streamlines audio production by automatically eliminating breath sounds.
Nidula Elgiriyewithana, N. D. Kodikara
― 6 min read
This project tests lane-following methods for safer self-driving vehicle operation.
Beñat Froemming-Aldanondo, Tatiana Rastoskueva, Michael Evans
― 4 min read
MM-DPCNs improve video analysis efficiency by learning features without labels.
Wenqian Xue, Chi Ding, Jose Principe
― 4 min read
MobileUNETR offers improved skin cancer detection through advanced image segmentation.
Shehan Perera, Yunus Erzurumlu, Deepak Gulati
― 6 min read
A new method enhances 3D image quality using dense metric depth.
Arkadeep Narayan Chaudhury, Igor Vasiljevic, Sergey Zakharov
― 6 min read
A new framework enhances the representation of neural fields on triangle meshes.
Avigail Cohen Rimon, Tal Shnitzer, Mirela Ben Chen
― 6 min read
A method to improve action recognition with fewer labeled videos and more unlabeled data.
Owais Iqbal, Omprakash Chakraborty, Aftab Hussain
― 6 min read
LongLLaVA improves multi-image understanding for various applications.
Xidong Wang, Dingjie Song, Shunian Chen
― 5 min read
This article examines how combining real and synthetic images boosts face recognition accuracy and fairness.
Andrea Atzori, Pietro Cosseddu, Gianni Fenu
― 5 min read
New method generates realistic 3D human models from single images using advanced video techniques.
Zhibin Liu, Haoye Dong, Aviral Chharia
― 5 min read
Innovative method combines machine learning and physics for solving differential equations.
Kai-liang Lu, Yu-meng Su, Zhuo Bi
― 6 min read
Introducing new metrics for assessing handwritten text generation systems.
Konstantina Nikolaidou, George Retsinas, Giorgos Sfikas
― 6 min read
A new method improves predictions of hand movements in videos for robots and virtual reality.
Junyi Ma, Xieyuanli Chen, Wentao Bao
― 5 min read
A new model enhances fashion item suggestions using geometry and visual data.
Ryotaro Shimizu, Yu Wang, Masanari Kimura
― 4 min read
A new method predicts BMI using handwriting styles and deep learning.
N. T. Diba, N. Akter, S. A. H. Chowdhury
― 6 min read
Study reveals voice data's role in recognizing emotions in Spanish speakers.
Elena Ortega-Beltrán, Josep Cabacas-Maso, Ismael Benito-Altamirano
― 5 min read
This study presents a model that integrates context for better facial expression recognition.
Florian Blume, Runfeng Qu, Pia Bideau
― 8 min read
New models improve road damage detection with drones, enhancing city safety.
Weichao Pan, Xu Wang, Wenqing Huan
― 5 min read
StyleTokenizer improves image generation by separating style and text instructions.
Wen Li, Muyuan Fang, Cheng Zou
― 7 min read
This approach combines autoencoders and diffusion techniques for clearer images.
Vighnesh Birodkar, Gabriel Barcik, James Lyon
― 6 min read
Plane2Depth improves depth estimation in complex scenes, addressing challenges of low texture.
Li Liu, Ruijie Zhu, Jiacheng Deng
― 6 min read
This research enhances depth estimation in robots using meta-learning for better performance in varied environments.
Cho-Ying Wu, Yiqi Zhong, Junying Wang
― 5 min read
A system helps identify Korean dishes for those with dietary needs.
Hoang Khanh Lam, Kahandakanaththage Maduni Pramuditha Perera
― 6 min read
New video generation method improves realism for self-driving car training.
Jianbiao Mei, Xuemeng Yang, Licheng Wen
― 6 min read
A new framework enhances text descriptions using images and structured data.
Tahsina Hashem, Weiqing Wang, Derry Tanti Wijaya
― 5 min read
FODA-PG enhances report generation from medical images for better diagnosis.
Kai Shu, Yuzhuo Jia, Ziyang Zhang
― 5 min read
A new method and dataset for automated cell analysis in brain research.
Valentina Vadori, Jean-Marie Graïc, Antonella Peruffo
― 4 min read
A new approach to create synthetic images efficiently for dataset distillation.
Zhe Li, Weitong Zhang, Sarah Cechnicka
― 8 min read
This project explores AI methods for more efficient garbage classification.
Jenil Kanani
― 5 min read
This study examines the use of generative systems for managing historical photographs in Catalan archives.
Èric Śanchez, Adrià Molina, Oriol Ramos Terrades
― 6 min read
MVTN improves hand gesture recognition through innovative multiscale techniques.
Mallika Garg, Debashis Ghosh, Pyari Mohan Pradhan
― 5 min read
This study assesses various visual models for understanding complex 3D scenes.
Yunze Man, Shuhong Zheng, Zhipeng Bao
― 8 min read
This study evaluates machine learning models for detecting trash in rivers.
Marga Don, Stijn Pinson, Blanca Guillen Cebrian
― 5 min read
GIMDiffusion simplifies 3D generation from text descriptions using geometry images.
Slava Elizarov, Ciara Rowles, Simon Donné
― 6 min read
RealisHuman enhances image quality by refining human features in generated images.
Benzhi Wang, Jingkai Zhou, Jingqi Bai
― 5 min read
A new method improves surface reconstruction from sparse images, ensuring detail and efficiency.
Rui Peng, Shihe Shen, Kaiqiang Xiong
― 6 min read
SegTalker enhances talking face videos with realistic textures and easy editing.
Lingyu Xiong, Xize Cheng, Jintao Tan
― 5 min read
TCDiff enhances synthetic face creation for better face recognition.
Bernardo Biesseck, Pedro Vidal, Luiz Coelho
― 5 min read
A new method for assessing robustness in ML classifiers using adversarial distance.
Georg Siedel, Ekagra Gupta, Andrey Morozov
― 6 min read