mlsanityData to 14 Oct 2026

SISMA: Semantic Face Image Synthesis with Mamba

arXiv 2509.17651 ↗First posted 22 Sep 2025cs.CVcited by 0

On the map: Diffusion models → Image generation, text-to-image and editing

also in 1 more
Abstract (arXiv)

Diffusion Models have become very popular for Semantic Image Synthesis (SIS) of human faces. Nevertheless, their training and inference is computationally expensive and their computational requirements are high due to the quadratic complexity of attention layers. In this paper, we propose a novel architecture called SISMA, based on the recently proposed Mamba. SISMA generates high quality samples by controlling their shape using a semantic mask at a reduced computational demand. We validated our approach through comprehensive experiments with CelebAMask-HQ, revealing that our architecture not only achieves a better FID score yet also operates at three times the speed of state-of-the-art architectures. This indicates that the proposed design is a viable, lightweight substitute to transformer-based models.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 33 of its 36 references that are in the map. First posted on arXiv 22 Sep 2025.

Builds on

33 of 36 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024Controllable Face Synthesis with Semantic Latent Diffusion Models19 Mar
  2. 2023T2I-Adapter: Learning Adapters to Dig out More Controllable Ability for Text-to-Image Diffusion Models16 Feb
  3. 2023Adding Conditional Control to Text-to-Image Diffusion Models10 Feb

Other references · 30

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2022Scalable Diffusion Models with Transformerscited by 6,054
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  4. 2022Flow Matching for Generative Modelingcited by 5,702
  5. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
  6. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  7. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,465
  8. 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,336
  9. 2019Semantic Image Synthesis with Spatially-Adaptive Normalizationcited by 1,299
  10. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,179
  11. 2019MaskGAN: Towards Diverse and Interactive Facial Image Manipulationcited by 628
  12. 2021Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layerscited by 608
18 more
  1. 2020In-Domain GAN Inversion for Real Image Editingcited by 420
  2. 2023Composer: Creative and Controllable Image Synthesis with Composable Conditionscited by 304
  3. 2019SEAN: Image Synthesis with Semantic Region-Adaptive Normalizationcited by 253
  4. 2022Semantic Image Synthesis via Diffusion Modelscited by 132
  5. 2019Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesiscited by 108
  6. 2024ZigMa: A DiT-style Zigzag Mamba Diffusion Modelcited by 84
  7. 2020Semantically Multi-modal Image Synthesiscited by 58
  8. 2020Efficient Semantic Image Synthesis via Class-Adaptive Normalizationcited by 58
  9. 2021SemanticStyleGAN: Learning Compositional Generative Priors for Controllable Image Synthesis and Editingcited by 58
  10. 2021Diverse Semantic Image Synthesis via Probability Distribution Modelingcited by 49
  11. 2021Image Synthesis via Semantic Compositioncited by 45
  12. 2020Dual Attention GANs for Semantic Image Synthesiscited by 41
  13. 2024Matten: Video Generation with Mamba-Attentioncited by 24
  14. 2023Automatic Generation of Semantic Parts for Face Image Synthesiscited by 5
  15. 2023Semantic Image Synthesis via Class-Adaptive Cross-Attentioncited by 5
  16. 2024Mamba-ST: State Space Model for Efficient Style Transfercited by 5
  17. 2025U-Shape Mamba: State Space Model for faster diffusioncited by 5
  18. 2024Adversarial Identity Injection for Semantic Face Image Synthesiscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024MaskMamba: A Hybrid Mamba-Transformer Model for Masked Image Generation
  2. 2025Mamba-CNN: A Hybrid Architecture for Efficient and Accurate Facial Beauty Prediction
  3. 2024I2I-Mamba: Multi-modal medical image synthesis via selective state space modeling
  4. 2026MMFace-DiT: A Dual-Stream Diffusion Transformer for High-Fidelity Multimodal Face Generation
  5. 2024DiM: Diffusion Mamba for Efficient High-Resolution Image Synthesis
  6. 2024CapHuman: Capture Your Moments in Parallel Universes
  7. 2025SynergyNet: Fusing Generative Priors and State-Space Models for Facial Beauty Prediction
  8. 2025Bringing Diversity from Diffusion Models to Semantic-Guided Face Asset Generation