NegROI: Click-Centric Uncertainty-Guided Refinement with Scene-Conditioned Negative Prompts for Robust Interactive 3D Segmentation

• URL: http://arxiv.org/abs/2607.05955v1
• Date: Tue, 07 Jul 2026 07:54:52 GMT
• Title: NegROI: Click-Centric Uncertainty-Guided Refinement with Scene-Conditioned Negative Prompts for Robust Interactive 3D Segmentation
• Abstract summary: Interactive 3D segmentation aims to extract object masks in point clouds with minimal user clicks.<n>Existing approaches still struggle with (i) coarse voxel resolution that blurs fine boundaries under limited clicks and (ii) hard false positives caused by confusing background structures.<n>We propose NegROI -- a novel transformer-based interactive framework that couples click-centric multi-resolution refinement with scene-conditioned negative prompts.<n>Our experiments on common benchmark datasets demonstrate improved click efficiency and reduced false positives, with stronger cross-dataset robustness than the state-of-the-art baselines.
• Score: 28.716564476160958
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Interactive 3D segmentation aims to extract object masks in point clouds with minimal user clicks. Despite recent progress, most existing approaches still struggle with (i) coarse voxel resolution that blurs fine boundaries under limited clicks and (ii) hard false positives caused by confusing background structures. These issues are exacerbated by density and scale shifts across datasets (e.g., dense RGB-D reconstructions vs. sparse LiDAR scans), where fixed refinement heuristics and purely click-driven decoding generalize poorly. To address them, we propose NegROI -- a novel transformer-based interactive framework that couples click-centric multi-resolution refinement with scene-conditioned negative prompts. Given a coarse voxel prediction, it refines only a local Region Of Interest (ROI) around the current click on a finer grid and fuses refined logits back to the coarse mask. To improve robustness and efficiency, we introduce uncertainty-driven selective refinement that prioritizes ambiguous regions. Meanwhile, we model hard background patterns via a set of scene-conditioned negative prompts obtained by cross-attention over scene tokens. We further stabilize these prompts with a diversity regularizer. Finally, we propose boundary-aware hard negative mining to supervise negative-prompt attention toward boundary-proximal, high-confidence false positives. Our experiments on common benchmark datasets (i.e., ScanNet, S3DIS, and KITTI) demonstrate improved click efficiency and reduced false positives, with stronger cross-dataset robustness than the state-of-the-art baselines.

Related papers

• U-CFR: Uncertainty-Guided Cascade Forward Refinement for Interactive Segmentation [2.3488056916440856]
  Uncertainty-Guided Cascade Forward Refinement (U-CFR)<n>We propose a novel inference-time framework that enables models to autonomously self-correct after each user interaction.<n>It reduces the required clicks by over 10% on challenging datasets like Berkeley.
  arXiv  Detail & Related papers  (2026-07-22T20:23:00Z)
• Visual Geometry Transformer in the Wild: Distractor-Free 3D Reconstruction [77.75911117889915]
  Current end-to-end multi-view 3D reconstruction methods achieve impressive results, but rely on a restrictive static assumption.<n>This reliance on idealized inputs causes even the most advanced methods to fail in real-world settings.<n>We propose Visual Geometry Transformer in the Wild, an end-to-end framework for robust reconstruction from inconsistent views.
  arXiv  Detail & Related papers  (2026-06-22T02:52:22Z)
• Non-Contrastive Vision-Language Learning with Predictive Embedding Alignment [12.336161969869567]
  We introduce NOVA, a NOn-contrastive Vision-language Alignment framework based on joint embedding prediction with distributional regularization.<n>We evaluate NOVA on zeroshot chest X-ray classification using ClinicalBERT as the text encoder and Vision Transformers trained from scratch on MIMIC-CXR.<n>Our results demonstrate that non-contrastive vision-language pretraining offers a simpler, more stable, and more effective alternative to contrastive methods.
  arXiv  Detail & Related papers  (2026-01-31T10:57:46Z)
• Extremal Contours: Gradient-driven contours for compact visual attribution [5.6220652636435915]
  We show how a star-supervised framework can achieve higher complexity with dense masks.<n>On ImageNets, it matches the extremal contours of dense masks while producing compact regions with improved run-to-run vision.
  arXiv  Detail & Related papers  (2025-11-03T10:02:21Z)
• PVNet: Point-Voxel Interaction LiDAR Scene Upsampling Via Diffusion Models [57.02789948234898]
  We propose PVNet, a diffusion model-based point-voxel interaction framework to perform LiDAR point cloud upsampling without dense supervision.<n>Specifically, we employ a sparse point cloud as the guiding condition and the synthesized point clouds derived from its nearby frames as the input.<n>In addition, we propose a point-voxel interaction module to integrate features from both points and voxels, which efficiently improves the environmental perception capability of each upsampled point.
  arXiv  Detail & Related papers  (2025-08-23T14:55:03Z)
• Capturing Fine-Grained Alignments Improves 3D Affordance Detection [0.46040036610482665]
  We propose LM-AD, a novel method for affordance detection in 3D point clouds.<n>We also introduce the Affordance Query Module (AQM), which efficiently captures fine-grained alignment between point clouds and text.<n>Our method outperformed existing approaches in terms of accuracy and mean Intersection over Union on the 3D AffordanceNet dataset.
  arXiv  Detail & Related papers  (2025-06-24T04:58:55Z)
• Hard-Label Black-Box Attacks on 3D Point Clouds [66.52447238776482]
  We introduce a novel 3D attack method based on a new spectrum-aware decision boundary algorithm to generate high-quality adversarial samples.<n>Experiments demonstrate that our attack competitively outperforms existing white/black-box attackers in terms of attack performance and adversary quality.
  arXiv  Detail & Related papers  (2024-11-30T09:05:02Z)
• SIRST-5K: Exploring Massive Negatives Synthesis with Self-supervised Learning for Robust Infrared Small Target Detection [53.19618419772467]
  Single-frame infrared small target (SIRST) detection aims to recognize small targets from clutter backgrounds. With the development of Transformer, the scale of SIRST models is constantly increasing. With a rich diversity of infrared small target data, our algorithm significantly improves the model performance and convergence speed.
  arXiv  Detail & Related papers  (2024-03-08T16:14:54Z)
• Implicit and Efficient Point Cloud Completion for 3D Single Object Tracking [9.372859423951349]
  We introduce two novel modules, i.e., Adaptive Refine Prediction (ARP) and Target Knowledge Transfer (TKT) Our model achieves state-of-the-art performance while maintaining a lower computational consumption.
  arXiv  Detail & Related papers  (2022-09-01T15:11:06Z)
• Progressive Coordinate Transforms for Monocular 3D Object Detection [52.00071336733109]
  We propose a novel and lightweight approach, dubbed em Progressive Coordinate Transforms (PCT) to facilitate learning coordinate representations. In this paper, we propose a novel and lightweight approach, dubbed em Progressive Coordinate Transforms (PCT) to facilitate learning coordinate representations.
  arXiv  Detail & Related papers  (2021-08-12T15:22:33Z)
• Learning to Estimate Hidden Motions with Global Motion Aggregation [71.12650817490318]
  Occlusions pose a significant challenge to optical flow algorithms that rely on local evidences. We introduce a global motion aggregation module to find long-range dependencies between pixels in the first image. We demonstrate that the optical flow estimates in the occluded regions can be significantly improved without damaging the performance in non-occluded regions.
  arXiv  Detail & Related papers  (2021-04-06T10:32:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.