MELT: Improve Composed Image Retrieval via the Modification Frequentation-Rarity Balance Network
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911556572282880 |
|---|---|
| author | Qiu, Guozhi Chen, Zhiwei Li, Zixu Huang, Qinlei Fu, Zhiheng Song, Xuemeng Hu, Yupeng |
| author_facet | Qiu, Guozhi Chen, Zhiwei Li, Zixu Huang, Qinlei Fu, Zhiheng Song, Xuemeng Hu, Yupeng |
| contents | Composed Image Retrieval (CIR) uses a reference image and a modification text as a query to retrieve a target image satisfying the requirement of ``modifying the reference image according to the text instructions''. However, existing CIR methods face two limitations: (1) frequency bias leading to ``Rare Sample Neglect'', and (2) susceptibility of similarity scores to interference from hard negative samples and noise. To address these limitations, we confront two key challenges: asymmetric rare semantic localization and robust similarity estimation under hard negative samples. To solve these challenges, we propose the Modification frEquentation-rarity baLance neTwork MELT. MELT assigns increased attention to rare modification semantics in multimodal contexts while applying diffusion-based denoising to hard negative samples with high similarity scores, enhancing multimodal fusion and matching. Extensive experiments on two CIR benchmarks validate the superior performance of MELT. Codes are available at https://github.com/luckylittlezhi/MELT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_29291 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MELT: Improve Composed Image Retrieval via the Modification Frequentation-Rarity Balance Network Qiu, Guozhi Chen, Zhiwei Li, Zixu Huang, Qinlei Fu, Zhiheng Song, Xuemeng Hu, Yupeng Computer Vision and Pattern Recognition Artificial Intelligence Composed Image Retrieval (CIR) uses a reference image and a modification text as a query to retrieve a target image satisfying the requirement of ``modifying the reference image according to the text instructions''. However, existing CIR methods face two limitations: (1) frequency bias leading to ``Rare Sample Neglect'', and (2) susceptibility of similarity scores to interference from hard negative samples and noise. To address these limitations, we confront two key challenges: asymmetric rare semantic localization and robust similarity estimation under hard negative samples. To solve these challenges, we propose the Modification frEquentation-rarity baLance neTwork MELT. MELT assigns increased attention to rare modification semantics in multimodal contexts while applying diffusion-based denoising to hard negative samples with high similarity scores, enhancing multimodal fusion and matching. Extensive experiments on two CIR benchmarks validate the superior performance of MELT. Codes are available at https://github.com/luckylittlezhi/MELT. |
| title | MELT: Improve Composed Image Retrieval via the Modification Frequentation-Rarity Balance Network |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2603.29291 |