Research map / 3D vision / 3D generation
Content-Aware Warping for View Synthesis
Mantang Guo, Junhui Hou, Jing Jin, Hui Liu, Huanqiang Zeng, Jiwen Lu
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 |
|---|---|---|---|
| add content embedding (ablation) | base view synthesis model · DTU (2 inputs) | average PSNR (dB): 20.69 -> 21.85 | Improved |
| add global embedding (ablation) | content-aware warping model · DTU (2 inputs) | average PSNR (dB): 21.85 -> 21.94 | Improved |
| add adaptive PSV fusion (ablation) | content-aware warping model · DTU (2 inputs) | average PSNR (dB): 21.94 -> 22.15 | Improved |
| add feature-space warping (ablation) | content-aware warping model · DTU (2 inputs) | average PSNR (dB): 22.15 -> 22.30 | Improved |
| add weight-smoothness loss (ablation) | content-aware warping model · DTU (2 inputs) | average PSNR (dB): 22.30 -> 22.41 | Improved |
| replace disparity-based warping with content-aware warping | Baseline (Warp) · Inria Sparse LF | average PSNR (dB): 28.27 -> 31.00 | Improved |
| add more accurate disparity input to Jin et al. | Jin et al. · Inria Sparse LF | average PSNR (dB): 27.54 -> 29.00 | Improved |
| use new components over preliminary version | Guo et al. · Inria Sparse LF | average PSNR (dB): 30.17 -> 31.00 | Improved |
| use new components over preliminary version | Guo et al. · DTU | average PSNR (dB): 20.29 -> 22.41 | Improved |
| increase source views from 2 to 3 | content-aware warping model · DTU | average PSNR (dB): 22.41 -> 23.27 | Improved |
| increase source views from 3 to 4 | content-aware warping model · DTU | average PSNR (dB): 23.27 -> 23.72 | Improved |
| content-aware warping vs IBRNet and SVNVS on texture-less scenes | content-aware warping model · DTU scan66/67/106 | PSNR: scan66 23.57 vs IBRNet 25.09 | Hurt |
Closest papers by meaning
- Learning Dynamic Interpolation for Extremely Sparse Light Fields With Wide Baselines (ICCV 2021 · similarity 0.87)
- Context-Aware Synthesis for Video Frame Interpolation (CVPR 2018 · similarity 0.87)
- WarpGAN: Warping-Guided 3D GAN Inversion with Style-Based Novel View Inpainting (NeurIPS 2025 · similarity 0.85)
- Novel View Synthesis with Pixel-Space Diffusion Models (CVPR 2025 · similarity 0.85)
- GenWarp: Single Image to Novel Views with Semantic-Preserving Generative Warping (NeurIPS 2024 · similarity 0.84)
- Geometry-guided Online 3D Video Synthesis with Multi-View Temporal Consistency (CVPR 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.
