Hiroshi Murase
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Hiroshi Murase has a total of 85 co-authors. Here are the top co-authors by number of shared publications:
- Daisuke Deguchi
Nagoya University, Nagoya, Japan
20 shared publications - Jialei Chen
Graduate School of Informatics, Nagoya University, Nagoya, Japan
12 shared publications - Yasutomo Kawanishi
Guardian Robot Project R-IH, RIKEN
9 shared publications - Chenkai Zhang
Graduate School of Informatics, Nagoya University, Nagoya, Japan
7 shared publications - Ichiro Ide
Nagoya University, Nagoya, Japan
6 shared publications - Takatsugu Hirayama
Nagoya University
6 shared publications
Latest Publications
- Open Access
- 28 July 2026
- Published by Elsevier in SSRN Electronic Journal
Beyond Frame-level Representation for Zero-shot Video Action Recognition
Zero-shot video action recognition aims to recognize unseen action categories by transferring knowledge from seen categories through vision-language alignment. However, existing CLIP-based methods inherit image-centric representations, limiting their ability to capture hierarchical temporal dependencies and coherent action semantics across video sequences. To address this limitation, we propose HTM-CLIP, a unified vision-language framework that enhances frozen CLIP through hierarchical temporal modeling and conditional semantic enhancement. Specifically, hierarchical temporal modeling captures short-, mid-, and long-range dependencies in a progressive manner, enabling efficient video representation learning without global attention. Meanwhile, conditional semantic enhancement facilitates early visual-semantic interaction and category-aware feature refinement via gated prompt modulation. Together, these components learn temporally coherent and semantically discriminative video representations, improving vision-language alignment and zero-shot generalization. Extensive experiments demonstrate that HTM-CLIP consistently outperforms state-of-the-art methods while maintaining competitive computational efficiency.
- 28 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessBeyond Frame-level Representation for Zero-shot Video Action Recognition
Zero-shot video action recognition aims to recognize unseen action categories by transferring knowledge from seen categories through vision-language alignment. However, existing CLIP-based methods inherit image-centric representations, limiting their ability to capture hierarchical temporal dependencies and coherent action semantics across video sequences. To address this limitation, we propose HTM-CLIP, a unified vision-language framework that enhances frozen CLIP through hierarchical temporal modeling and conditional semantic enhancement. Specifically, hierarchical temporal modeling captures short-, mid-, and long-range dependencies in a progressive manner, enabling efficient video representation learning without global attention. Meanwhile, conditional semantic enhancement facilitates early visual-semantic interaction and category-aware feature refinement via gated prompt modulation. Together, these components learn temporally coherent and semantically discriminative video representations, improving vision-language alignment and zero-shot generalization. Extensive experiments demonstrate that HTM-CLIP consistently outperforms state-of-the-art methods while maintaining competitive computational efficiency.
- Open Access
- 25 July 2026
- Published by Elsevier in SSRN Electronic Journal
Not All Pixels Are Equal: Confidence-GuidedAttention for Feature Matching
Semi-dense feature matching methods have been significantly advanced by leveraging attention mechanisms to extract discriminative descriptors. However, most existing approaches treat all pixels equally during attention computations, which can potentially introduce noise and redundancy from irrelevant regions. To address this issue, we propose a Confidence-guided semi-dense Matcher (CgMatcher) that adaptively prunes attention weights for each pixel based on precomputed matching confidence maps. These maps are generated by evaluating the mutual similarity between feature pairs extracted from the backbone, where high confidence indicates a high potential for matching. Then the attention is refined through two steps: (1) a confidence-guided bias is introduced to adaptively adjust the attention distributions for each query pixel, avoiding irrelevant interactions between non-overlap pixels; (2) the corresponding confidence map is additionally employed to rescale value features during feature aggregation, attenuating the influence of uncertain regions. Moreover, a classification loss is introduced to encourage the backbone's features to discriminate between matchable and non-matchable regions. Extensive experiments on three benchmarks demonstrate that the proposal outperforms existing state-of-the-art methods.
- 25 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessNot All Pixels Are Equal: Confidence-GuidedAttention for Feature Matching
Semi-dense feature matching methods have been significantly advanced by leveraging attention mechanisms to extract discriminative descriptors. However, most existing approaches treat all pixels equally during attention computations, which can potentially introduce noise and redundancy from irrelevant regions. To address this issue, we propose a Confidence-guided semi-dense Matcher (CgMatcher) that adaptively prunes attention weights for each pixel based on precomputed matching confidence maps. These maps are generated by evaluating the mutual similarity between feature pairs extracted from the backbone, where high confidence indicates a high potential for matching. Then the attention is refined through two steps: (1) a confidence-guided bias is introduced to adaptively adjust the attention distributions for each query pixel, avoiding irrelevant interactions between non-overlap pixels; (2) the corresponding confidence map is additionally employed to rescale value features during feature aggregation, attenuating the influence of uncertain regions. Moreover, a classification loss is introduced to encourage the backbone's features to discriminate between matchable and non-matchable regions. Extensive experiments on three benchmarks demonstrate that the proposal outperforms existing state-of-the-art methods.
- Open Access
- 20 July 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (8) , 353
Correction: Training-Free Open-Vocabulary Semantic Segmentation with Context Pyramid Refinement
- 20 July 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (8) , 353
Open AccessCorrection: Training-Free Open-Vocabulary Semantic Segmentation with Context Pyramid Refinement
- Open Access
- 6 June 2026
- Published by Elsevier in SSRN Electronic Journal
LCS-VPR: Landmark-aware Contextual queries with Skipping re-ranking mechanism for Visual Place Recognition
Visual Place Recognition (VPR) aims to identify previously visited locations by matching query images against a geo-tagged reference database. Despite significant progress, existing methods suffer from two core limitations. (i) Granularity misalignment: prevailing approaches either aggregate dense pixel features into global descriptors, capturing fine-grained textures but treating all regions equally without awareness of high-level scene structure, or leverage semantic segmentation for scene-level understanding while discarding discriminative low-level details. Neither paradigm jointly exploits pixel-level cues and segment-level semantics. (ii) Cross-image blindness: each image is encoded independently, without considering which visual patterns persist across different views of the same place and which are merely transient distractors such as lighting changes or moving objects. We propose LCS-VPR (Landmark-aware Cross-image Semantic framework for Visual Place Recognition). To resolve granularity misalignment, landmark-aware semantic queries selectively attend to specific landmark regions, fusing scene semantics and fine-grained spatial details into a unified representation, while a query matching loss supervises these queries to encode place-discriminative semantics. To resolve cross-image blindness, a cross-image encoder module operates across images, allowing the model to compare visual patterns and identify co-occurring structures shared among same-place images while suppressing appearance-specific noise that appears in only individual views, thereby directly resolving cross-image blindness. Furthermore, an entropy-guided confidence estimation mechanism quantifies retrieval reliability and bypasses unnecessary re-ranking, improving inference efficiency. To the best of our knowledge, LCS-VPR is the first VPR framework to jointly integrate landmark-level semantics and pixel-level details into a unified pipeline. Extensive experiments demonstrate that LCS-VPR achieves state-of-the-art performance.
- 6 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessLCS-VPR: Landmark-aware Contextual queries with Skipping re-ranking mechanism for Visual Place Recognition
Visual Place Recognition (VPR) aims to identify previously visited locations by matching query images against a geo-tagged reference database. Despite significant progress, existing methods suffer from two core limitations. (i) Granularity misalignment: prevailing approaches either aggregate dense pixel features into global descriptors, capturing fine-grained textures but treating all regions equally without awareness of high-level scene structure, or leverage semantic segmentation for scene-level understanding while discarding discriminative low-level details. Neither paradigm jointly exploits pixel-level cues and segment-level semantics. (ii) Cross-image blindness: each image is encoded independently, without considering which visual patterns persist across different views of the same place and which are merely transient distractors such as lighting changes or moving objects. We propose LCS-VPR (Landmark-aware Cross-image Semantic framework for Visual Place Recognition). To resolve granularity misalignment, landmark-aware semantic queries selectively attend to specific landmark regions, fusing scene semantics and fine-grained spatial details into a unified representation, while a query matching loss supervises these queries to encode place-discriminative semantics. To resolve cross-image blindness, a cross-image encoder module operates across images, allowing the model to compare visual patterns and identify co-occurring structures shared among same-place images while suppressing appearance-specific noise that appears in only individual views, thereby directly resolving cross-image blindness. Furthermore, an entropy-guided confidence estimation mechanism quantifies retrieval reliability and bypasses unnecessary re-ranking, improving inference efficiency. To the best of our knowledge, LCS-VPR is the first VPR framework to jointly integrate landmark-level semantics and pixel-level details into a unified pipeline. Extensive experiments demonstrate that LCS-VPR achieves state-of-the-art performance.
- 23 May 2026
- Published by Springer Nature
- p. 16-29
Multi-Group Vision Semantic Centroid for Semantic Segmentation
- 23 May 2026
- Published by Springer Nature
- p. 16-29
Multi-Group Vision Semantic Centroid for Semantic Segmentation
- 23 May 2026
- Published by Springer Nature
- p. 30-42
A Cross-Modal Knowledge Distillation Approach for RGB-to-Infrared Video Action Recognition
- 23 May 2026
- Published by Springer Nature
- p. 30-42
A Cross-Modal Knowledge Distillation Approach for RGB-to-Infrared Video Action Recognition
- 21 May 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (6) , 281
Training-Free Open-Vocabulary Semantic Segmentation with Context Pyramid Refinement
- 21 May 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (6) , 281
Training-Free Open-Vocabulary Semantic Segmentation with Context Pyramid Refinement
- Open Access
- 24 March 2026
- Published by Elsevier in SSRN Electronic Journal
Multiscale Temporal Modeling with Semantic Enhancement for Zero-shot Video Action Recognition
Zero-shot video action recognition relies on effective alignment between visual and semantic features to recognize actions from unseen categories. However, existing CLIP-based video methods typically rely on frame-level representations with limited cross-frame interaction, making it difficult to maintain cross-frame consistency and leading to wrong predictions. To address this, we propose MTM-CLIP, a framework for multiscale temporal modeling. It explicitly captures temporal action patterns across short-term, mid-term, and global-term windows in an adaptive manner. In addition, we introduce a conditional semantic enhancement mechanism, where learnable gated prompts are integrated into the CLIP visual encoder to modulate cross-frame temporal features. These features are refined via semantic guidance to enhance discriminative features at key temporal positions. By jointly modeling cross-frame temporal interactions and conditional semantic modulation, MTM-CLIP improves vision-language alignment and zero-shot generalization. Extensive experiments show that MTM-CLIP significantly outperforms prior CLIP-based methods across multiple zero-shot video action recognition benchmarks.
- 24 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMultiscale Temporal Modeling with Semantic Enhancement for Zero-shot Video Action Recognition
Zero-shot video action recognition relies on effective alignment between visual and semantic features to recognize actions from unseen categories. However, existing CLIP-based video methods typically rely on frame-level representations with limited cross-frame interaction, making it difficult to maintain cross-frame consistency and leading to wrong predictions. To address this, we propose MTM-CLIP, a framework for multiscale temporal modeling. It explicitly captures temporal action patterns across short-term, mid-term, and global-term windows in an adaptive manner. In addition, we introduce a conditional semantic enhancement mechanism, where learnable gated prompts are integrated into the CLIP visual encoder to modulate cross-frame temporal features. These features are refined via semantic guidance to enhance discriminative features at key temporal positions. By jointly modeling cross-frame temporal interactions and conditional semantic modulation, MTM-CLIP improves vision-language alignment and zero-shot generalization. Extensive experiments show that MTM-CLIP significantly outperforms prior CLIP-based methods across multiple zero-shot video action recognition benchmarks.
- 16 February 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (3) , 122
Semantic-Centric Alignment for Zero-shot Panoptic Segmentation with Limited Data
- 16 February 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (3) , 122
Semantic-Centric Alignment for Zero-shot Panoptic Segmentation with Limited Data
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 1-12
BiXFormer: A Robust Framework for Maximizing Modality Effectiveness in Multi-Modal Semantic Segmentation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 1-12
Open AccessBiXFormer: A Robust Framework for Maximizing Modality Effectiveness in Multi-Modal Semantic Segmentation