BMVA Symposium on AI for Visual Arts

Vision methods built around photorealistic data struggle with the abstraction, stylisation and narrative structure of paintings, comics, illustration and sculpture, while generative models raise open questions of controllability, provenance, authorship and alignment with artistic intent. Topics include generative models for art and design, sketch-based interfaces, AI for animation and visual effects, perception of non-photorealistic content, computer vision for cultural heritage digitisation and restoration, vision-language models for art interpretation, neural rendering and 3D Gaussian splatting, human-AI co-creation, provenance and authenticity verification, style transfer and computational art analysis, and the ethics of AI-generated art. Chaired by Deblina Bhattacharjee, the meeting features keynotes including Yi-Zhe Song and Neill Campbell, contributed presentations, poster and demo sessions and a panel discussion. The event is in person only; registration costs £30 including lunch and refreshments.