arXiv Computer Vision October 21, 2026 By Wenbin Liao, Hao Zhu, Yike Ma, Hao Jiang, Feng Dai

Efficient SAM3 Adaptation for Multi-Class Semantic Segmentation via Dense Competitive Representations

Read the original on arXiv Computer Vision →

The paper introduces DCR‑SAM, a framework that adapts the Segment Anything Model 3 (SAM3) for multi‑class semantic segmentation. By decoupling cross‑modal interactions, aligning textual concepts with visual representations, and adding a dense competition head and learnable background token, DCR‑SAM reduces inference latency, mitigates semantic misclassifications, and suppresses overlapping predictions. Experiments show 2.3%–10.7% performance gains, a 33× speedup over native SAM3, and strong robustness across benchmarks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Hugging Face Trending Papers

5d ago

Efficient SAM3 Adaptation for Multi-Class Semantic Segmentation via Dense Competitive Representations

The Segment Anything Model 3 (SAM3) has achieved significant progress in Promptable Concept Segmentation (PCS) by processing short noun phrases to generate segmentation masks with unique instance iden...

computer-vision efficiency multimodal benchmarks

arXiv AI

Oct 15

UnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity

UnSAMv2 is a self‑supervised extension of the Segment Anything Model that can segment objects at any granularity without human annotations. It discovers mask‑granularity pairs and introduces a granularity control embedding for continuous scale adjustment, requiring only 6K unlabeled images and 0.02% extra parameters. The model outperforms SAM/SAM‑2 across 11+ benchmarks, improving metrics such as 1‑IoU from 58.0 to 73.1 and achieving state‑of‑the‑art performance on language‑guided segmentation.

By Junwei Yu, Trevor Darrell, XuDong Wang

llms rag computer-vision multimodal benchmarks

arXiv AI

Sep 30

ActiveSAM: Fast and Accurate Open-Vocabulary Semantic Segmentation with Frozen SAM 3

arXiv:2606.16996v2 Announce Type: replace-cross Abstract: Segment Anything Model 3 (SAM 3) provides a strong frozen backbone for concept-prompted segmentation, but applying it directly to open-vocabu...

By Tran Dinh Tien, Zhiqiang Shen

agents computer-vision robotics benchmarks

arXiv Computer Vision

Sep 28

ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models

ProCAP introduces a probabilistic cross-attentive prompt learning framework for vision-language models like CLIP, enabling improved cross-modal interaction without updating the backbone. It jointly learns visual and textual prompt tokens, linking them via stacked bidirectional multi-head cross-attention to refine each branch across prompt depth. The method incorporates Gaussian parameterization of prompt tokens, lightweight KL and L2 regularization, and a compact symmetric InfoNCE head to align image features with class-level text representations, achieving strong few-shot base-to-novel performance and competitive transfer results across multiple datasets and benchmarks.

By Hiwa Azeez Abbas, Fatemeh Daneshfar, Moloud Abdar

llms multimodal benchmarks

arXiv Computer Vision

Aug 25

Sa2VA: Marrying SAM2 with MLLM for Dense Grounded Understanding of Images and Videos

arXiv:2501.04001v4 Announce Type: replace Abstract: This work presents Sa2VA, the first comprehensive, unified model for dense grounded understanding of both images and videos. Unlike existing multi-...

By Haobo Yuan, Xiangtai Li, Tao Zhang, Yueyi Sun, Zilong Huang, Shilin Xu, Shunping Ji, Yunhai Tong, Lu Qi, Jiashi Feng, Ming-Hsuan Yang

llms computer-vision fine-tuning multimodal benchmarks

arXiv Computer Vision

Sep 30

Reprogramming Vision-Language Models via Structured Prompt Reparameterization

arXiv:2609.36680v1 Announce Type: new Abstract: Visual reprogramming adapts pretrained models to downstream tasks by modifying their input and output interfaces while keeping the backbone fixed. In v...

By Zizhao Li, Chengyi Cai, Mohammed Yaqoob Ansari, Feng Liu, Joseph West, Kourosh Khoshelham

llms rag multimodal benchmarks

One email a morning, machine-written

Related stories

Efficient SAM3 Adaptation for Multi-Class Semantic Segmentation via Dense Competitive Representations

UnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity

ActiveSAM: Fast and Accurate Open-Vocabulary Semantic Segmentation with Frozen SAM 3

ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models

Sa2VA: Marrying SAM2 with MLLM for Dense Grounded Understanding of Images and Videos

Reprogramming Vision-Language Models via Structured Prompt Reparameterization