Deep Cross-Modal Projection Learning for Image-Text Matching
By Ying Zhang, Huchuan Lu. European Conference on Computer Vision, 2018
473 citations recorded by the source corpus.
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/1a86eb42952412ee02e3f6da06f874f1946eff6b
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Ling Shao
- Xiaogang Wang
- Dacheng Tao
- Luc Van Gool
- Ming-Hsuan Yang
- Jianbing Shen
- Steven C. H. Hoi
- Hongsheng Li
Citations in the corpus
- CAMP: Cross-Modal Adaptive Message Passing for Text-Image Retrieval (2019, methodology): “[33] used projection classification loss which categorized the vector projection of representations from one modality onto another with the improved normsoftmax loss.”
- Multi-Modality Cross Attention Network for Image and Sentence Matching (2020, unknown): “4 CMPM (ResNet) [54] 49.”
- Graph Structured Network for Image-Text Matching (2020, background): “The main goal of global correspondence learning methods [19, 4, 29, 29, 31, 22, 34, 1] is to maximize similarity of matched image-text pairs.”
- Adversarial Text-to-Image Synthesis: A Review (2021, methodology): “In addition to unconditional and conditional losses in the discriminator, the cross-modal projection matching (CMPM) and cross-modal projection classification (CMPC) losses [82] are used to align input captions with generated images.”
- Deep Learning for Person Re-Identification: A Survey and Outlook (2020, methodology): “A cross projection learning method is also presented in [186] to learn a shared space with image-to-text matching.”
- Scalable Deep Multimodal Learning for Cross-Modal Retrieval (2019, methodology): “…problems, inspired by the great success of deep neural networks (DNN) in representation learning [18], several DNN-based approaches have been proposed to learn the complex nonlinear transformations for cross-modal retrieval in an unsupervised [2, 6, 40] or supervised [9, 10, 30, 38, 44] manner.”
- Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval (2023, methodology): “To demonstrate the effectiveness of our proposed similarity distribution matching (SDM) loss, we compare it with the commonly used cross-modal projection matching (CMPM) loss [53] (No.”