Research map / Image generation / Text-to-image diffusion
ChArtist: Generating Pictorial Charts with Unified Spatial and Subject Control
Shishi Xiao, Tongyu Zhou, David H. Laidlaw, Gromit Yeuk-Yin Chan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use skeleton control with ChArtist instead of Canny/depth ControlNet for text alignment | ChArtist · Spatial control only, bar chart | CLIP-T: 0.249 -> 0.304 | Improved |
| use ChArtist instead of ControlNet-Depth for line chart data accuracy | ChArtist · Spatial control only, line chart | Data Acc: 0.858 -> 0.920 | Improved |
| use ChArtist instead of InPainting for bar chart data accuracy | ChArtist · Spatial control only, bar chart | Data Acc: 0.923 -> 0.894 | Hurt |
| use ChArtist instead of SDEdit for pie chart data accuracy | ChArtist · Spatial control only, pie chart | Data Acc: 0.836 -> 0.778 | Hurt |
| use dual control with Spatially-Gated Attention instead of ControlNet-Canny + IP-Adapter on line charts | ChArtist · Dual control, line chart | Data Acc: 0.728 -> 0.905 | Improved |
| use ChArtist instead of Paint-by-Example for bar charts under dual control | ChArtist · Dual control, bar chart | DINO: 0.586 -> 0.837 | Improved |
| use ChArtist instead of Qwen-Image-Edit on pie charts under dual control | ChArtist · Dual control, pie chart | Data Acc: 0.765 -> 0.753 | Hurt |
| use ChArtist instead of GPT-Img-1 for CLIP-IQA | ChArtist · Dual control, image quality | CLIP-IQA: 0.712 -> 0.657 | Hurt |
| lower beta from 0.9 to 0.3 in Spatially-Gated Attention (ablation) | ChArtist · Line chart dual control | Data Acc: 0.729 -> 0.927 | Improved |
| lower beta from 0.9 to 0.3 in Spatially-Gated Attention (ablation) | ChArtist · Line chart dual control | DINO: 0.775 -> 0.732 | Hurt |
| lower beta from 0.6 to 0.3 in Spatially-Gated Attention (ablation) | ChArtist · Line chart dual control | Data Acc: 0.876 -> 0.927 | Improved |
| human study: rank ChArtist against baselines | ChArtist · Online user study (300 participants) | average rank: 2.957 (2nd) spatial, 2.845 (2nd) subject | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SDEdit — SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations (ICLR 2022) · better on 2 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding (CVPR 2026 · similarity 0.86)
- ChartE$^{3}$: A Comprehensive Benchmark for End-to-End Chart Editing (ICML 2026 · similarity 0.86)
- Factuality Matters: When Image Generation and Editing Meet Structured Visuals (ICLR 2026 · similarity 0.85)
- ChartGalaxy: A Dataset for Infographic Chart Understanding and Generation (ICLR 2026 · similarity 0.85)
- Visual Self-Refine: A Pixel-Guided Paradigm for Accurate Chart Parsing (ICLR 2026 · similarity 0.84)
- ChartSketcher: Reasoning with Multimodal Feedback and Reflection for Chart Understanding (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
