Post
57
šš”š New Research Alert - ICCV 2025 (Oral)! ššŖš
š Title: LoftUp: Learning a Coordinate-based Feature Upsampler for Vision Foundation Models š
š Description: LoftUp is a coordinate-based transformer that upscales the low-resolution features of VFMs (e.g. DINOv2 and CLIP) using cross-attention and self-distilled pseudo-ground truth (pseudo-GT) from SAM.
š„ Authors: Haiwen Huang, Anpei Chen, Volodymyr Havrylov, Andreas Geiger, and Dan Zhang
š Conference: ICCV, 19 ā 23 Oct, 2025 | Honolulu, Hawai'i, USA šŗšø
š Paper: LoftUp: Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models (2504.14032)
š Github Page: https://andrehuang.github.io/loftup-site
š Repository: https://github.com/andrehuang/loftup
š ICCV-2023-25-Papers: https://github.com/DmitryRyumin/ICCV-2023-25-Papers
š Added to the Foundation Models and Representation Learning Section: https://github.com/DmitryRyumin/ICCV-2023-25-Papers/blob/main/sections/2025/main/foundation-models-and-representation-learning.md
š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin
š Keywords: #LoftUp #VisionFoundationModels #FeatureUpsampling #Cross-AttentionTransformer #CoordinateBasedLearning #SelfDistillation #PseudoGroundTruth #RepresentationLearning #AI #ICCV2025 #ResearchHighlight
š Title: LoftUp: Learning a Coordinate-based Feature Upsampler for Vision Foundation Models š
š Description: LoftUp is a coordinate-based transformer that upscales the low-resolution features of VFMs (e.g. DINOv2 and CLIP) using cross-attention and self-distilled pseudo-ground truth (pseudo-GT) from SAM.
š„ Authors: Haiwen Huang, Anpei Chen, Volodymyr Havrylov, Andreas Geiger, and Dan Zhang
š Conference: ICCV, 19 ā 23 Oct, 2025 | Honolulu, Hawai'i, USA šŗšø
š Paper: LoftUp: Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models (2504.14032)
š Github Page: https://andrehuang.github.io/loftup-site
š Repository: https://github.com/andrehuang/loftup
š ICCV-2023-25-Papers: https://github.com/DmitryRyumin/ICCV-2023-25-Papers
š Added to the Foundation Models and Representation Learning Section: https://github.com/DmitryRyumin/ICCV-2023-25-Papers/blob/main/sections/2025/main/foundation-models-and-representation-learning.md
š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin
š Keywords: #LoftUp #VisionFoundationModels #FeatureUpsampling #Cross-AttentionTransformer #CoordinateBasedLearning #SelfDistillation #PseudoGroundTruth #RepresentationLearning #AI #ICCV2025 #ResearchHighlight
