Pith. sign in

REVIEW 4 major objections 3 minor 42 references

Tri-FusionNet: Enhancing Image Description Generation with Transformer-based Fusion Network and Dual Attention Mechanism

T0 review · 4 major / 3 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Tri-FusionNet claims record image captioning scores by fusing ViT, RoBERTa, and CLIP with dual attention.

desk verdict Architecture is a plausible mashup of ViT, RoBERTa, and CLIP, but the reported results are internally inconsistent and the SOTA claims do not hold up. read the letter →

arxiv 2504.16761 v1 pith:DD46DNLM submitted 2025-04-23 cs.CV

Lakshita Agarwal , Bindu Verma This is my paper · ORCID
classification cs.CV
keywords ImageDescriptionGenerationNaturalLanguageProcessingComputerVisionTransformersDualAttentionMechanismTransformerRoBERTaCLIP
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that image captioning can be improved by fusing three transformer-based modules: a Vision Transformer encoder that uses dual attention over spatial windows and channel groups, a RoBERTa-based decoder, and a CLIP integration module that aligns visual and textual embeddings with contrastive learning. On the MS-COCO, Flickr30k, and Flickr8k benchmarks, it reports scores that would surpass every previous model, including a BLEU-1 of 0.893 and CIDEr of 1.880 on MS-COCO. If those numbers are correct, the architecture is a new high-water mark for accessible, contextually rich image description generation.

What carries the argument

The central object is the dual-attention mechanism inside the ViT encoder. It runs spatial window self-attention over local, non-overlapping image regions in parallel with channel-group self-attention across the channel dimension, and concatenates the two outputs; this gives linear complexity in both spatial and channel dimensions while capturing local and global context. The RoBERTa decoder generates the caption text from word embeddings and the ViT output, and the CLIP integration module concatenates the two embeddings and aligns them with a contrastive loss so that corresponding image and text pairs sit close in a shared space. These three modules are the machinery that carries the argument from raw images to the reported metric improvements.

What would settle it

Re-run the test split for MS-COCO with the published metric implementations and confirm that Tri-FusionNet yields BLEU-1 0.893, BLEU-4 0.725, CIDEr 1.880, METEOR 0.780, and ROUGE-L 0.689; if the reproduced numbers fall well below the best published models (e.g., CIDEr below 1.4), the state-of-the-art claim fails.

Watch

Extended reading notes

Core claim

Tri-FusionNet integrates three pretrained transformer families: a Vision Transformer encoder with a dual attention mechanism that combines spatial window attention and channel-group attention, a RoBERTa decoder that contextualizes text embeddings, and a CLIP integration module that maps visual and textual embeddings into a shared space via contrastive loss. The paper reports that this fusion, optimized with Adam and cross-entropy loss, achieves state-of-the-art results across all tested metrics on MS-COCO (BLEU-1 0.893, BLEU-4 0.725, CIDEr 1.880, METEOR 0.780, ROUGE-L 0.689), Flickr30k, and Flickr8k. The central discovery claimed is that dual-attention ViT encoding combined with contrastive CLIP alignment and RoBERTa decoding produces descriptions that are both more semantically accurate and more fluent than those of earlier transformer captioning models.

Load-bearing premise

The reported scores were obtained under the standard image-captioning evaluation protocol on the correct test splits and are correctly assigned to the specified model and metric columns.

Editorial extensions

If this is right

  • If the reported scores are reproducible, fusing a dual-attention ViT encoder with a pretrained language decoder and CLIP alignment becomes a viable template for reaching state-of-the-art image captioning.
  • The dual-attention encoder's linear complexity suggests the architecture can scale to higher-resolution images or longer caption sequences without quadratic cost.
  • The CLIP contrastive alignment may reduce the semantic gap between visual and textual representations, making generated captions more grounded in image content.
  • The model's success on three different benchmarks implies the fusion recipe transfers across datasets of different sizes and styles.
  • The qualitative failures reported (e.g., misidentifying a motorcycle as a car) indicate the remaining bottleneck is fine-grained object recognition, pointing to where future gains could come from.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: if the dual-attention and contrastive components are responsible for the gains, the same recipe could be transferred to adjacent tasks such as visual question answering or text-to-image generation, where aligning visual and textual spaces is also central.
  • The paper's own tables contain inconsistencies (the abstract's Flickr8k values do not match Table VIII), so an independent re-evaluation on a fixed protocol would be needed before treating the exact record numbers as settled.
  • A testable extension would be ablating CLIP alignment while keeping the dual-attention encoder to isolate how much of the CIDEr gain actually comes from contrastive fusion rather than from the encoder alone.
  • Because the model concatenates ViT and RoBERTa embeddings and passes them through dense layers, the architecture could likely be trimmed to smaller pretrained backbones for deployment on edge devices, at some cost in accuracy.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 3 minor

Summary. The paper proposes Tri-FusionNet, an image-captioning model that combines a Vision Transformer (ViT) encoder with a dual-attention mechanism, a RoBERTa decoder, and a CLIP integration module. The authors report state-of-the-art results on MS-COCO, Flickr30k, and Flickr8k across BLEU, METEOR, CIDEr, and ROUGE-L, and present ablations, qualitative examples, and comparisons with prior work. The central claim is that Tri-FusionNet surpasses all existing models on the three benchmarks, with headline numbers of BLEU 0.893, METEOR 0.780, CIDEr 1.880, and ROUGE-L 0.689 on MS-COCO.

Significance. If the reported results were reliable, the paper would be a substantial contribution: a single architecture outperforming all prior image-captioning systems on three standard benchmarks by a wide margin would be noteworthy, and the fusion of ViT, RoBERTa, and CLIP with dual attention is a plausible design direction worth exploring. The paper's qualitative examples and heatmap analysis do provide some illustrative value. However, the significance hinges entirely on the correctness of the quantitative evaluation, and the manuscript's internal inconsistencies and improbable metric values undermine that foundation. The paper does not ship code, reproducible evaluation scripts, or the promised supplementary files, so there is no independent verification path either.

major comments (4)
  1. [Abstract vs. Table VIII and Section IV.B] The reported numbers are internally inconsistent. The abstract states Flickr8k results as CIDEr 1.483, METEOR 0.358, ROUGE-L 0.789, while Table VIII reports the same dataset as CIDEr 1.381, METEOR 0.389, ROUGE-L 0.654. Moreover, Section IV.B's prose assigns the abstract's Flickr8k triplet (1.483, 0.358, 0.789) to MS-COCO instead of the values in Table VIII (1.88, 0.78, 0.689). Since the central state-of-the-art claim rests on these tables, this mutual contradiction means the evaluation evidence is not trustworthy as presented.
  2. [Section IV.A, Tables III and IV] The prose misattributes every metric in the ViT+RoBERTa+CLIP rows. For Table III (Flickr30k), the row contains B-1=0.741, B-2=0.621, B-3=0.573, B-4=0.428, C=1.092, M=0.389, R-L=0.432, but the text states '0.741 for METEOR, 0.621 for ROUGE-L, 0.573 for BLEU-4, 0.428 for BLEU-3, 1.092 for CIDEr, 0.389 for BLEU-2, and 0.432 for BLEU-1.' The same misattribution occurs in the Flickr8k discussion. If the prose cannot correctly map reported numbers to their columns, the reader has no way to know which values actually correspond to which metrics.
  3. [Section IV.C, Table X] The claimed MS-COCO METEOR of 0.780 and CIDEr of 1.880 are far outside the range of all comparable systems in Table X, where the best listed METEOR is 0.392 and the best CIDEr is 1.415. No evaluation protocol is given: the paper does not specify the test split, beam size, caption preprocessing, or whether the online MS-COCO test server was used for the numbers in Table VIII. The sentence in Section IV.C that Tri-FusionNet 'breaks all other records as the new baseline' is therefore unsupported by reproducible evidence.
  4. [Section III.A.3 and Supplementary Material] The architecture description is incomplete in a way that prevents reproducibility. RoBERTa is a masked-language-model encoder, yet the paper uses it as a decoder without specifying the output layer, vocabulary projection, or how captions are generated during inference. The implementation details are repeatedly deferred to 'supplementary.tex' and 'supplementary.txt' (e.g., Sections III.A.1, III.A.4, and IV), but no supplementary file is included with the manuscript. This makes it impossible to verify the proposed method or the reported numbers.
minor comments (3)
  1. [Section IV.A] The text before Table III says the table reports Flickr8k results, while the caption and the actual values correspond to Flickr30k; please correct the mismatch between the caption, the prose, and the table content.
  2. [Tables X-XIII] The 'BLEU' column is ambiguous; it should state whether it refers to BLEU-1, BLEU-4, or an average, because the comparison with prior work is otherwise not meaningful.
  3. [Section IV.B] The paragraph after Table VIII repeats the wrong metric-to-dataset assignment: BLEU values for MS-COCO are followed by CIDEr/METEOR/ROUGE-L values 1.483/0.358/0.789, which are neither the MS-COCO values in Table VIII nor the Flickr8k values in Table VIII.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: empirical training/evaluation with no fitted-as-predicted step; self-citation [2] is not load-bearing.

full rationale

The paper's central claim is empirical ('Tri-FusionNet breaks all other records as the new baseline', Section IV.C), and the chain from model definition to scores is a trained neural net, not an equation-level derivation. The attention equations (1)-(5) are standard self-attention and window/channel-group formulations; Eq. (6) and (7) simply name the RoBERTa decoder and a concatenation. No parameter is fitted to a metric and then reported as a prediction; the reported BLEU/METEOR/CIDEr/Rouge-L values (Table VIII, X) are the outputs of training, so no 'prediction' is equivalent by construction to an input. The only self-citation is the authors' own survey [2] in the Introduction, and it is not load-bearing for the architecture or results. I therefore find no circular derivation. Two serious non-circular flaws are flagged per the reviewing rule: (i) the paper repeatedly defers essential details to a supplementary file that is not included ('The process of patch extraction and positional encoding is thoroughly explained in the supplementary.tex file' — Section III.A.1; 'The architectural explanation of the CLIP integration module is explained in the supplementary.txt file' — Section III.A.4; 'The implementation details of the proposed framework is demonstrated thoroughly in supplementary.tex file' — Section IV; 'The file supplementary.txt is available alongside this manuscript' — Supplementary Material), so the implementation is not self-contained; (ii) the reported scores are internally inconsistent (abstract Flickr8k CIDEr 1.483/METEOR 0.358/Rouge-L 0.789 vs Table VIII 1.381/0.389/0.654, and Section IV.A prose misattributes Table III values). These undermine verification of the SOTA claim but are not circularity. Score 1.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new postulates or entities beyond the architecture itself. The load-bearing support comes from pretrained models and standard equations; the key missing piece is the unspecified set of hand-chosen hyperparameters and the unstated generative mechanism.

free parameters (1)
  • Unspecified architecture hyperparameters = not reported
    Patch size, number of attention heads Nh, number of spatial windows Nw, channel groups Ng, and dense-layer dimensions are chosen by hand but never stated; the claimed performance depends on these choices, and their absence blocks replication.
assumptions (3)
  • standard math Standard self-attention equations (1)-(4) are valid and applicable to image patches.
    The paper uses conventional softmax dot-product attention from the transformer literature, though it does not cite the original Transformer paper.
  • domain assumption Pre-trained ViT, RoBERTa, and CLIP weights can be transferred and fine-tuned for caption generation.
    The model depends on pretrained representations being available and compatible; no training from scratch is described.
  • ad hoc to paper RoBERTa, a masked language model, can serve as a caption decoder that generates descriptions via the unspecified output layers.
    RoBERTa is not an autoregressive generator; the paper asserts the RoBERTa decoder produces descriptions (Eq. 6) without specifying a generative head, autoregressive loop, or inference procedure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Tri-FusionNet: Enhancing Image Description Generation with Transformer-based Fusion Network and Dual Attention Mechanism." pith.science (2026). https://pith.science/paper/DD46DNLM

@misc{pith2026250416761,
  author       = {Pith},
  title        = {Pith review of: Tri-FusionNet: Enhancing Image Description Generation with Transformer-based Fusion Network and Dual Attention Mechanism},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DD46DNLM}},
  note         = {Machine review of arXiv:2504.16761}
}
read the original abstract

Image description generation is essential for accessibility and AI understanding of visual content. Recent advancements in deep learning have significantly improved natural language processing and computer vision. In this work, we propose Tri-FusionNet, a novel image description generation model that integrates transformer modules: a Vision Transformer (ViT) encoder module with dual-attention mechanism, a Robustly Optimized BERT Approach (RoBERTa) decoder module, and a Contrastive Language-Image Pre-Training (CLIP) integrating module. The ViT encoder, enhanced with dual attention, focuses on relevant spatial regions and linguistic context, improving image feature extraction. The RoBERTa decoder is employed to generate precise textual descriptions. CLIP's integrating module aligns visual and textual data through contrastive learning, ensuring effective combination of both modalities. This fusion of ViT, RoBERTa, and CLIP, along with dual attention, enables the model to produce more accurate, contextually rich, and flexible descriptions. The proposed framework demonstrated competitive performance on the Flickr30k and Flickr8k datasets, with BLEU scores ranging from 0.767 to 0.456 and 0.784 to 0.479, CIDEr scores of 1.679 and 1.483, METEOR scores of 0.478 and 0.358, and ROUGE-L scores of 0.567 and 0.789, respectively. On MS-COCO, the framework obtained BLEU scores of 0.893 (B-1), 0.821 (B-2), 0.794 (B-3), and 0.725 (B-4). The results demonstrate the effectiveness of Tri-FusionNet in generating high-quality image descriptions.

Figures

Figures reproduced from arXiv: 2504.16761 by the authors.

Figure 1
Figure 1. Structural representation of Tri-FusionNet framework: The architecture consists of three phases: firstly, high-level visual features are first extracted [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Example of obtained heat maps based on dual-attention mechanism. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Graphical representation of the results obtained from the datasets for [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 29 canonical work pages

  1. [1]

    Deep learning approaches on image captioning: A review,

    T. Ghandi, H. Pourreza, and H. Mahyar, “Deep learning approaches on image captioning: A review,” ACM Computing Surveys , vol. 56, no. 3, pp. 1–39, 2023

    2023

  2. [2]

    From methods to datasets: A survey on image-caption generators,

    L. Agarwal and B. Verma, “From methods to datasets: A survey on image-caption generators,” Multimedia Tools and Applications , vol. 83, no. 9, pp. 28 077–28 123, 2024

    work page 2024

  3. [3]

    A survey on vision transformer,

    K. Han, Y . Wang, H. Chen, X. Chen, J. Guo, Z. Liu, Y . Tang, A. Xiao, C. Xu, Y . Xuet al., “A survey on vision transformer,” IEEE transactions on pattern analysis and machine intelligence, vol. 45, no. 1, pp. 87–110, 2022

    2022

  4. [4]

    Text augmentation using bert for image captioning,

    V . Atliha and D. ˇSeˇsok, “Text augmentation using bert for image captioning,” Applied Sciences, vol. 10, no. 17, p. 5978, 2020

    work page 2020

  5. [5]

    Contrastive language- image pre-training with knowledge graphs,

    X. Pan, T. Ye, D. Han, S. Song, and G. Huang, “Contrastive language- image pre-training with knowledge graphs,” Advances in Neural Infor- mation Processing Systems , vol. 35, pp. 22 895–22 910, 2022

    work page 2022

  6. [6]

    Entangled transformer for image captioning,

    G. Li, L. Zhu, P. Liu, and Y . Yang, “Entangled transformer for image captioning,” in Proceedings of the IEEE/CVF international conference on computer vision , 2019, pp. 8928–8937

    work page 2019

  7. [7]

    Meshed-memory transformer for image captioning,

    M. Cornia, M. Stefanini, L. Baraldi, and R. Cucchiara, “Meshed-memory transformer for image captioning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2020, pp. 10 578–10 587

    2020

  8. [8]

    Multimodal transformer with multi- view visual representation for image captioning,

    J. Yu, J. Li, Z. Yu, and Q. Huang, “Multimodal transformer with multi- view visual representation for image captioning,” IEEE transactions on circuits and systems for video technology , vol. 30, no. 12, pp. 4467– 4480, 2019

    2019

Show all 42 references
  1. [9]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark et al., “Learning transferable visual models from natural language supervision,” in International conference on machine learning . PMLR, 2021, pp. 8748–8763

    2021

  2. [10]

    Minigpt-v2: large language model as a unified interface for vision-language multi-task learning,

    J. Chen, D. Zhu, X. Shen, X. Li, Z. Liu, P. Zhang, R. Krishnamoorthi, V . Chandra, Y . Xiong, and M. Elhoseiny, “Minigpt-v2: large language model as a unified interface for vision-language multi-task learning,” arXiv preprint arXiv:2310.09478 , 2023

    arXiv

  3. [11]

    Samt- generator: A second-attention for image captioning based on multi-stage transformer network,

    X. Yang, Y . Yang, S. Ma, Z. Li, W. Dong, and M. Wo ´zniak, “Samt- generator: A second-attention for image captioning based on multi-stage transformer network,” Neurocomputing, vol. 593, p. 127823, 2024

    2024

  4. [12]

    S2 transformer for image captioning

    P. Zeng, H. Zhang, J. Song, and L. Gao, “S2 transformer for image captioning.” in IJCAI, 2022, pp. 1608–1614

    2022

  5. [13]

    Ca-captioner: A novel concentrated attention for image captioning,

    X. Yang, Y . Yang, J. Wu, W. Sun, S. Ma, and Z. Hou, “Ca-captioner: A novel concentrated attention for image captioning,” Expert Systems with Applications, vol. 250, p. 123847, 2024

    2024

  6. [14]

    Image cap- tioning using transformer-based double attention network,

    H. Parvin, A. R. Naghsh-Nilchi, and H. M. Mohammadi, “Image cap- tioning using transformer-based double attention network,” Engineering Applications of Artificial Intelligence , vol. 125, p. 106545, 2023

    2023

  7. [15]

    With a little help from your own past: Prototypical memory networks for image captioning,

    M. Barraco, S. Sarto, M. Cornia, L. Baraldi, and R. Cucchiara, “With a little help from your own past: Prototypical memory networks for image captioning,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 3021–3031

    2023

  8. [16]

    Haav: Hierarchical aggregation of augmented views for image captioning,

    C.-W. Kuo and Z. Kira, “Haav: Hierarchical aggregation of augmented views for image captioning,” in Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition , 2023, pp. 11 039– 11 049

    2023

  9. [17]

    Dual vision transformer,

    T. Yao, Y . Li, Y . Pan, Y . Wang, X.-P. Zhang, and T. Mei, “Dual vision transformer,” IEEE transactions on pattern analysis and machine intelligence, 2023

    2023

  10. [18]

    Spt: Spatial pyramid transformer for image captioning,

    H. Zhang, P. Zeng, L. Gao, X. Lyu, J. Song, and H. T. Shen, “Spt: Spatial pyramid transformer for image captioning,” IEEE Transactions on Circuits and Systems for Video Technology , 2023

    2023

  11. [19]

    Show and tell: A neural image caption generator,

    O. Vinyals, A. Toshev, S. Bengio, and D. Erhan, “Show and tell: A neural image caption generator,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2015, pp. 3156–3164

    2015

  12. [20]

    Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation,

    J. Li, D. Li, C. Xiong, and S. Hoi, “Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation,” in International conference on machine learning . PMLR, 2022, pp. 12 888–12 900

    2022

  13. [21]

    A topic-based multi-channel attention model under hybrid mode for image caption,

    K. Qian and L. Tian, “A topic-based multi-channel attention model under hybrid mode for image caption,” Neural Computing and Applications , vol. 34, no. 3, pp. 2207–2216, 2022

    2022

  14. [22]

    Transformer model incorporating local graph semantic attention for image caption,

    K. Qian, Y . Pan, H. Xu, and L. Tian, “Transformer model incorporating local graph semantic attention for image caption,” The Visual Computer, pp. 1–12, 2023

    2023

  15. [23]

    Dynamic-balanced double-attention fusion for image captioning,

    C. Wang and X. Gu, “Dynamic-balanced double-attention fusion for image captioning,” Engineering Applications of Artificial Intelligence , vol. 114, p. 105194, 2022

    2022

  16. [24]

    Improving image captioning by leveraging intra-and inter-layer global representation in transformer network,

    J. Ji, Y . Luo, X. Sun, F. Chen, G. Luo, Y . Wu, Y . Gao, and R. Ji, “Improving image captioning by leveraging intra-and inter-layer global representation in transformer network,” in Proceedings of the AAAI conference on artificial intelligence , vol. 35, no. 2, 2021, pp. 1655– 1663

    2021

  17. [25]

    Geometry attention transformer with position-aware lstms for image captioning,

    C. Wang, Y . Shen, and L. Ji, “Geometry attention transformer with position-aware lstms for image captioning,” Expert systems with ap- plications, vol. 201, p. 117174, 2022

    2022

  18. [26]

    Vision-enhanced and consensus- aware transformer for image captioning,

    S. Cao, G. An, Z. Zheng, and Z. Wang, “Vision-enhanced and consensus- aware transformer for image captioning,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 10, pp. 7005–7018, 2022

    2022

  19. [27]

    A novel cross-fusion method of different types of features for image captioning,

    L. Lou, K. Lu, and J. Xue, “A novel cross-fusion method of different types of features for image captioning,” in 2023 International Joint Conference on Neural Networks (IJCNN) . IEEE, 2023, pp. 1–8

    2023

  20. [28]

    Transformer- based local-global guidance for image captioning,

    H. Parvin, A. R. Naghsh-Nilchi, and H. M. Mohammadi, “Transformer- based local-global guidance for image captioning,” Expert Systems with Applications, vol. 223, p. 119774, 2023

    2023

  21. [29]

    Deep captioning with multimodal recurrent neural networks (m-rnn),

    J. Mao, “Deep captioning with multimodal recurrent neural networks (m-rnn),” arXiv preprint arXiv:1412.6632 , 2014

    2014 arXiv

  22. [30]

    Re- view networks for caption generation,

    Z. Yang, Y . Yuan, Y . Wu, W. W. Cohen, and R. R. Salakhutdinov, “Re- view networks for caption generation,” Advances in neural information processing systems, vol. 29, 2016

    2016

  23. [31]

    Semantic compositional networks for visual captioning,

    Z. Gan, C. Gan, X. He, Y . Pu, K. Tran, J. Gao, L. Carin, and L. Deng, “Semantic compositional networks for visual captioning,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 5630–5639

    2017

  24. [32]

    Knowing when to look: Adaptive attention via a visual sentinel for image captioning,

    J. Lu, C. Xiong, D. Parikh, and R. Socher, “Knowing when to look: Adaptive attention via a visual sentinel for image captioning,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 375–383

    2017

  25. [33]

    Self- critical sequence training for image captioning,

    S. J. Rennie, E. Marcheret, Y . Mroueh, J. Ross, and V . Goel, “Self- critical sequence training for image captioning,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2017, pp. 7008–7024

    2017

  26. [34]

    Gatecap: Gated spatial and semantic attention model for image captioning,

    S. Wang, L. Lan, X. Zhang, and Z. Luo, “Gatecap: Gated spatial and semantic attention model for image captioning,” Multimedia Tools and Applications, vol. 79, pp. 11 531–11 549, 2020

    2020

  27. [35]

    Boosting image captioning with attributes,

    T. Yao, Y . Pan, Y . Li, Z. Qiu, and T. Mei, “Boosting image captioning with attributes,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 4894–4902

    2017

  28. [36]

    Bottom-up and top-down attention for image captioning and visual question answering,

    P. Anderson, X. He, C. Buehler, D. Teney, M. Johnson, S. Gould, and L. Zhang, “Bottom-up and top-down attention for image captioning and visual question answering,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2018, pp. 6077–6086

    2018

  29. [37]

    Recurrent fusion network for image captioning,

    W. Jiang, L. Ma, Y .-G. Jiang, W. Liu, and T. Zhang, “Recurrent fusion network for image captioning,” in Proceedings of the European conference on computer vision (ECCV) , 2018, pp. 499–515

    2018

  30. [38]

    Exploring visual relationship for image captioning,

    T. Yao, Y . Pan, Y . Li, and T. Mei, “Exploring visual relationship for image captioning,” in Proceedings of the European conference on computer vision (ECCV) , 2018, pp. 684–699

    2018

  31. [39]

    Auto-encoding scene graphs for image captioning,

    X. Yang, K. Tang, H. Zhang, and J. Cai, “Auto-encoding scene graphs for image captioning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2019, pp. 10 685–10 694

    2019

  32. [40]

    Attention on attention for image captioning,

    L. Huang, W. Wang, J. Chen, and X.-Y . Wei, “Attention on attention for image captioning,” in Proceedings of the IEEE/CVF international conference on computer vision , 2019, pp. 4634–4643

    2019

  33. [41]

    Image captioning using vision encoder decoder model,

    A. Abdelaal, N. F. ELshafey, N. W. Abdalah, N. H. Shaaban, S. A. Okasha, T. Yasser, M. Fathi, K. M. Fouad, and I. Abdelbaky, “Image captioning using vision encoder decoder model,” in 2024 International Conference on Machine Intelligence and Smart Innovation (ICMISI) . IEEE, 20...

    2024

  34. [42]

    Optimal trans- formers based image captioning using beam search,

    A. Shetty, Y . Kale, Y . Patil, R. Patil, and S. Sharma, “Optimal trans- formers based image captioning using beam search,” Multimedia Tools and Applications, vol. 83, no. 16, pp. 47 963–47 977, 2024

    2024

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.