OVS-DINO structurally aligns DINO with SAM to revitalize attenuated boundary features, achieving SOTA gains of 2.1% average and 6.3% on Cityscapes in weakly-supervised open-vocabulary segmentation.
Localizing objects with self-supervised transformers and no labels
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
fields
cs.CV 8representative citing papers
Register tokens enhance pixel-space DiT training and output quality via cleaner high-noise feature maps, and a dual-stream design adds further gains with little overhead.
RefCD enables unsupervised category-aware object detection by using feature similarity between predicted objects and unlabeled reference images to guide category learning.
TunnelMIND recalibrates language-guided defect proposals via dense visual consistency and reconstructs them into structured defect entities with attributes for severity grading and retrieval-grounded engineering reports, reporting F1 scores of 0.68, 0.78, and 0.72 on visible, GPR, and road defect任务.
ViCrop-Det uses spatial attention entropy from the decoder to dynamically crop and refine small-object regions in transformer detectors during inference.
ViTs exhibit lazy aggregation by relying on irrelevant background patches for global semantics, and selectively integrating patch features into the CLS token reduces this effect and improves results across label-, text-, and self-supervision.
Franca introduces nested Matryoshka clustering and positional disentanglement in a transparent SSL pipeline to deliver open-source vision models competitive with closed proprietary systems.
PANC augments Normalized Cut with anchor-augmented token graphs using priors to steer spectral partitions, yielding mIoU gains of 2.3-8.7% over baselines on DUTS-TE, DUT-OMRON, and CrackForest.
citing papers explorer
-
OVS-DINO: Open-Vocabulary Segmentation via Structure-Aligned SAM-DINO with Language Guidance
OVS-DINO structurally aligns DINO with SAM to revitalize attenuated boundary features, achieving SOTA gains of 2.1% average and 6.3% on Cityscapes in weakly-supervised open-vocabulary segmentation.
-
Registers Matter for Pixel-Space Diffusion Transformers
Register tokens enhance pixel-space DiT training and output quality via cleaner high-noise feature maps, and a dual-stream design adds further gains with little overhead.
-
Reference-based Category Discovery: Unsupervised Object Detection with Category Awareness
RefCD enables unsupervised category-aware object detection by using feature similarity between predicted objects and unlabeled reference images to guide category learning.
-
Training-Free Tunnel Defect Inspection and Engineering Interpretation via Visual Recalibration and Entity Reconstruction
TunnelMIND recalibrates language-guided defect proposals via dense visual consistency and reconstructs them into structured defect entities with attributes for severity grading and retrieval-grounded engineering reports, reporting F1 scores of 0.68, 0.78, and 0.72 on visible, GPR, and road defect任务.
-
ViCrop-Det: Spatial Attention Entropy Guided Cropping for Training-Free Small-Object Detection
ViCrop-Det uses spatial attention entropy from the decoder to dynamically crop and refine small-object regions in transformer detectors during inference.
-
Vision Transformers Need More Than Registers
ViTs exhibit lazy aggregation by relying on irrelevant background patches for global semantics, and selectively integrating patch features into the CLS token reduces this effect and improves results across label-, text-, and self-supervision.
-
Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning
Franca introduces nested Matryoshka clustering and positional disentanglement in a transparent SSL pipeline to deliver open-source vision models competitive with closed proprietary systems.
-
PANC: Prior-Aware Normalized Cut via Anchor-Augmented Token Graphs
PANC augments Normalized Cut with anchor-augmented token graphs using priors to steer spectral partitions, yielding mIoU gains of 2.3-8.7% over baselines on DUTS-TE, DUT-OMRON, and CrackForest.