FlashFPS accelerates FPS via candidate/iteration pruning and inter-layer caching, delivering 5.16x GPU speedup and 2.69x on accelerators with negligible accuracy loss.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
PointNTP serializes point clouds into geometry-ordered patch sequences and applies causal next-token prediction with stop-gradient targets for decoder-free self-supervised pre-training, reporting competitive results on ScanObjectNN, ShapeNetPart, and S3DIS.
Diff-SBSR uses a frozen Stable Diffusion backbone enhanced by multimodal CLIP and BLIP features plus Circle-T loss to outperform prior methods on zero-shot sketch-based 3D shape retrieval benchmarks.
MV-HGNN uses hierarchical graph convolutions on multi-view 3D features plus CLIP semantic prototypes to outperform prior methods on sketch-based 3D retrieval in both category-level and zero-shot settings.
Geometric Reward Credit Assignment disentangles rewards to geometric tokens and adds reprojection consistency to boost 3D keypoint accuracy from 0.64 to 0.93 and bounding box IoU to 0.686 on a ShapeNetCore benchmark while preserving 2D performance.
citing papers explorer
-
FlashFPS: Efficient Farthest Point Sampling for Large-Scale Point Clouds via Pruning and Caching
FlashFPS accelerates FPS via candidate/iteration pruning and inter-layer caching, delivering 5.16x GPU speedup and 2.69x on accelerators with negligible accuracy loss.
-
Rethinking Point Clouds as Sequences: A Causal Next-Token Predictive Learning Framework
PointNTP serializes point clouds into geometry-ordered patch sequences and applies causal next-token prediction with stop-gradient targets for decoder-free self-supervised pre-training, reporting competitive results on ScanObjectNN, ShapeNetPart, and S3DIS.
-
Diff-SBSR: Learning Multimodal Feature-Enhanced Diffusion Models for Zero-Shot Sketch-Based 3D Shape Retrieval
Diff-SBSR uses a frozen Stable Diffusion backbone enhanced by multimodal CLIP and BLIP features plus Circle-T loss to outperform prior methods on zero-shot sketch-based 3D shape retrieval benchmarks.
-
Multi-View Hierarchical Graph Neural Network for Sketch-Based 3D Shape Retrieval
MV-HGNN uses hierarchical graph convolutions on multi-view 3D features plus CLIP semantic prototypes to outperform prior methods on sketch-based 3D retrieval in both category-level and zero-shot settings.
-
Reinforcing 3D Understanding in Point-VLMs via Geometric Reward Credit Assignment
Geometric Reward Credit Assignment disentangles rewards to geometric tokens and adds reprojection consistency to boost 3D keypoint accuracy from 0.64 to 0.93 and bounding box IoU to 0.686 on a ShapeNetCore benchmark while preserving 2D performance.