SCI-CLIP: Segment-Centric Inference with Reference Memory for Training-Free Open-Vocabulary Segmentation

• URL: http://arxiv.org/abs/2608.05627v1
• Date: Thu, 06 Aug 2026 05:47:30 GMT
• Title: SCI-CLIP: Segment-Centric Inference with Reference Memory for Training-Free Open-Vocabulary Segmentation
• Abstract summary: We present SCI-CLIP, a segment-centric inference framework built around the principle that the same region abstraction should organize all stages of dense open-vocabulary prediction.<n> SCI-CLIP turns frozen CLIP-style features into spatially coherent, context-aware, and retrieval-compatible dense predictions without any training.
• Score: 0.6015898117103069
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Training-free open-vocabulary segmentation remains limited by a missing inference abstraction. Frozen vision-language features are produced at patch level, yet dense prediction requires a unit that simultaneously governs feature interaction, spatial support, contextual recovery, and retrieval-based correction. We present SCI-CLIP, a segment-centric inference framework built around the principle that the same region abstraction should organize all stages of dense open-vocabulary prediction. SCI-CLIP first induces a region-consistent interaction graph over frozen visual tokens, then reconstructs dense features by propagating values over this graph, augmenting them with selective cross-window support only where local evidence is insufficient. The same segment abstraction is subsequently used to construct and query an offline reference memory, aligning exemplar retrieval with the units on which prediction is made. SCI-CLIP turns frozen CLIP-style features into spatially coherent, context-aware, and retrieval-compatible dense predictions without any training. SCI-CLIP consistently improves the structural quality of dense predictions, the robustness of contextual reasoning, and the alignment of exemplar-based correction, yielding stronger open-vocabulary segmentation across eight benchmarks. Project code is available at: https://github.com/mzamini92/SCICLIP.

Related papers

• TraceCLIP: Recovering Local Semantics from Patch-to-CLS Contributions [20.430455131220544]
  We introduce TraceCLIP, a training-free framework that recovers latent patch-level semantic evidence.<n>We show that TraceCLIP achieves gains of 1.3 to 4.5 points in average mIoU over the strongest prior training-free methods.
  arXiv  Detail & Related papers  (2026-07-28T11:25:27Z)
• Fine-grained CLIP fine-tuning with self-annotated region alignment [57.28808560528696]
  Contrastive Language-Image Pre-training has been shown to have limitations in its fine-grained dense feature representation.<n>We propose SFF-CLIP (Self-annotated Fine-grained Fine-tuning for CLIP) to boost the fine-grained representation ability in the CLIP fine-tuning.
  arXiv  Detail & Related papers  (2026-07-15T10:06:39Z)
• Weave of Formal Thought [51.56484100374058]
  We introduce Weave of Formal Thought (WoFT), a paradigm uniting rigorous syntactic validation with learned structural representations.<n>We present a latent-variable fine-tuning method training the language model to interleave non-terminal grammar symbols directly into generation.<n>For Python, fine-tuning StarCoder2-3B with our RWS objective reduces per-token cross-entropy by 14.3% relative to a text-only SFT baseline.
  arXiv  Detail & Related papers  (2026-06-24T15:58:11Z)
• DouC: Dual-Branch CLIP for Training-Free Open-Vocabulary Segmentation [0.6015898117103069]
  DouC is a training-free dual-branch CLIP framework that decomposes dense prediction into two complementary components.<n> OG-CLIP improves patch-level reliability via lightweight, inference-time token gating.<n>FADE-CLIP injects external structural priors through proxy attention guided by frozen vision foundation models.
  arXiv  Detail & Related papers  (2026-04-27T20:59:01Z)
• Structure-Aware Feature Rectification with Region Adjacency Graphs for Training-Free Open-Vocabulary Semantic Segmentation [22.409969687852506]
  We propose a structure-aware feature rectification approach that incorporates instance-specific priors derived directly from the image.<n>Our method effectively suppresses segmentation noise, improves region-level consistency, and achieves strong performance on multiple open-vocabulary segmentation benchmarks.
  arXiv  Detail & Related papers  (2025-12-08T10:00:36Z)
• Generalized Decoupled Learning for Enhancing Open-Vocabulary Dense Perception [71.26728044621458]
  DeCLIP is a novel framework that enhances CLIP by decoupling the self-attention module to obtain content'' and context'' features respectively.<n>It consistently achieves state-of-the-art performance across a broad spectrum of tasks, including 2D detection and segmentation, 3D instance segmentation, video instance segmentation, and 6D object pose estimation.
  arXiv  Detail & Related papers  (2025-08-15T06:43:51Z)
• Partial CLIP is Enough: Chimera-Seg for Zero-shot Semantic Segmentation [55.486872677160015]
  We propose Chimera-Seg, which integrates a segmentation backbone as the body and a CLIP-based semantic head as the head.<n>Specifically, Chimera-Seg comprises a trainable segmentation model and a CLIP Semantic Head (CSH), which maps dense features into the CLIP-aligned space.<n>We also propose Selective Global Distillation (SGD), which distills knowledge from dense features exhibiting high similarity to the CLIP CLS token.
  arXiv  Detail & Related papers  (2025-06-27T09:26:50Z)
• DeCLIP: Decoupled Learning for Open-Vocabulary Dense Perception [21.87721909270275]
  DeCLIP is a novel framework that enhances CLIP with content'' and context'' features respectively.<n>It significantly outperforms existing methods across multiple open-vocabulary dense prediction tasks.
  arXiv  Detail & Related papers  (2025-05-07T13:46:34Z)
• Refining CLIP's Spatial Awareness: A Visual-Centric Perspective [10.936397225984107]
  Contrastive Language-Image Pre-training excels in global alignment with language but exhibits limited sensitivity to spatial information.<n>Recent approaches have introduced Region-Language Alignment to enhance CLIP's performance in dense multimodal tasks.<n>We propose the Spatial Correlation Distillation (SCD) framework, which preserves CLIP's inherent spatial structure and mitigates the above degradation.
  arXiv  Detail & Related papers  (2025-04-03T07:04:56Z)
• Open-Vocabulary Segmentation with Semantic-Assisted Calibration [68.41025728960176]
  We study open-vocabulary segmentation (OVS) through calibrating in-vocabulary and domain-biased embedding space with contextual prior of CLIP.<n>We present a Semantic-assisted CAlibration Network (SCAN) to achieve state-of-the-art performance on open-vocabulary segmentation benchmarks.
  arXiv  Detail & Related papers  (2023-12-07T07:00:09Z)
• DenseCLIP: Extract Free Dense Labels from CLIP [130.3830819077699]
  Contrastive Language-Image Pre-training (CLIP) has made a remarkable breakthrough in open-vocabulary zero-shot image recognition. DenseCLIP+ surpasses SOTA transductive zero-shot semantic segmentation methods by large margins. Our finding suggests that DenseCLIP can serve as a new reliable source of supervision for dense prediction tasks.
  arXiv  Detail & Related papers  (2021-12-02T09:23:01Z)

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.