The subspace intervention framework reveals that pre-training objectives shape how ViTs encode geometric information in compressible low-rank subspaces, with peak precision at intermediate layers.
The international journal of robotics research32(11), 1231–1237 (2013)
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CV 6years
2026 6verdicts
UNVERDICTED 6roles
dataset 1polarities
use dataset 1representative citing papers
Low-rank decoder adaptation enables efficient test-time optimization for zero-shot depth completion by updating only the subspace containing depth-relevant information.
ICDepth adapts text-to-video diffusion transformers for video depth estimation via in-context conditioning, achieving SOTA results on benchmarks with 6-13x less training data than prior generative methods.
RT-SFOD adapts dual-head detectors like YOLOv10 for source-free object detection via DHF pseudo-label fusion and MARD loss, delivering 1.4-3.5% mAP gains with 1.3x higher throughput and ~2x fewer parameters than prior SFOD methods.
Layer analysis of DINOv3 shows non-uniform 3D geometric knowledge concentrated in deeper layers, enabling a last-layer-centric recombination module that improves monocular depth estimation accuracy to state-of-the-art levels.
SS3D pretrains an end-to-end feed-forward 3D estimator on filtered YouTube-8M videos via SfM self-supervision, MVS filtering, and expert distillation, delivering stronger zero-shot transfer and fine-tuning than prior self-supervised baselines.
citing papers explorer
-
Understanding Geometric Representations in Self-Supervised Vision Transformers via Subspace Intervention
The subspace intervention framework reveals that pre-training objectives shape how ViTs encode geometric information in compressible low-rank subspaces, with peak precision at intermediate layers.
-
Efficient Test-Time Optimization for Depth Completion via Low-Rank Decoder Adaptation
Low-rank decoder adaptation enables efficient test-time optimization for zero-shot depth completion by updating only the subspace containing depth-relevant information.
-
ICDepth: Taming Video Diffusion Models for Video Depth Estimation via In-Context Conditioning
ICDepth adapts text-to-video diffusion transformers for video depth estimation via in-context conditioning, achieving SOTA results on benchmarks with 6-13x less training data than prior generative methods.
-
Real-Time Source-Free Object Detection
RT-SFOD adapts dual-head detectors like YOLOv10 for source-free object detection via DHF pseudo-label fusion and MARD loss, delivering 1.4-3.5% mAP gains with 1.3x higher throughput and ~2x fewer parameters than prior SFOD methods.
-
Last-Layer-Centric Feature Recombination: Unleashing 3D Geometric Knowledge in DINOv3 for Monocular Depth Estimation
Layer analysis of DINOv3 shows non-uniform 3D geometric knowledge concentrated in deeper layers, enabling a last-layer-centric recombination module that improves monocular depth estimation accuracy to state-of-the-art levels.
-
SS3D: End2End Self-Supervised 3D from Web Videos
SS3D pretrains an end-to-end feed-forward 3D estimator on filtered YouTube-8M videos via SfM self-supervision, MVS filtering, and expert distillation, delivering stronger zero-shot transfer and fine-tuning than prior self-supervised baselines.