A framework uses modality-agnostic prompts to adapt SAM for multi-modal camouflaged object detection, with a mask refine module for better boundaries.
SAM2-Adapter: Evaluating & adapting Seg- ment Anything 2 in downstream tasks: Camouflage, shadow, medical image segmentation, and more
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 6years
2026 6verdicts
UNVERDICTED 6roles
background 1polarities
background 1representative citing papers
PEPA is a post-encoder adapter combining target-conditioned snake upsampling and adaptive differentiable thresholding that improves topological metrics over region overlap when added to frozen-encoder curvilinear segmentation baselines.
M⁴-SAM equips SAM2 with modality-aware MoE-LoRA, gated multi-level fusion, and pseudo-guided initialization to reach state-of-the-art on RGB-D video salient object detection.
Permutation-COMQ is a new post-training quantization algorithm that reorders weights within layers and uses only dot-product and rounding steps to deliver the highest reported accuracy for 2-, 4-, and 8-bit medical foundation models.
DifferSeg introduces learnable differential operators for modality fusion and cross-frequency decoder interactions, claiming superior performance over 67 prior methods on 29 datasets across 18 tasks.
BED-SAM2 enhances the SAM2 vision model by integrating monocular geometric priors to improve boundary delineation in object segmentation tasks.
citing papers explorer
-
Modality-Agnostic Prompt Learning for Multi-Modal Camouflaged Object Detection
A framework uses modality-agnostic prompts to adapt SAM for multi-modal camouflaged object detection, with a mask refine module for better boundaries.
-
From Reconstruction to Decision: A Post-Encoder Plug-in Adapter for Curvilinear Segmentation
PEPA is a post-encoder adapter combining target-conditioned snake upsampling and adaptive differentiable thresholding that improves topological metrics over region overlap when added to frozen-encoder curvilinear segmentation baselines.
-
M$^4$-SAM: Multi-Modal Mixture-of-Experts with Memory-Augmented SAM for RGB-D Video Salient Object Detection
M⁴-SAM equips SAM2 with modality-aware MoE-LoRA, gated multi-level fusion, and pseudo-guided initialization to reach state-of-the-art on RGB-D video salient object detection.
-
Weight Group-wise Post-Training Quantization for Medical Foundation Model
Permutation-COMQ is a new post-training quantization algorithm that reorders weights within layers and uses only dot-product and rounding steps to deliver the highest reported accuracy for 2-, 4-, and 8-bit medical foundation models.
-
DifferSeg: Towards Diverse Multimodal Binary Segmentation via Differential Perception and Frequency Guidance
DifferSeg introduces learnable differential operators for modality fusion and cross-frequency decoder interactions, claiming superior performance over 67 prior methods on 29 datasets across 18 tasks.
-
BED-SAM2: Boundary-Enhanced-Depth SAM2 via Monocular Geometric Priors
BED-SAM2 enhances the SAM2 vision model by integrating monocular geometric priors to improve boundary delineation in object segmentation tasks.