Research map / 3D vision / 3D generation
FIERY: Future Instance Prediction in Bird's-Eye View From Surround Monocular Cameras
Anthony Hu, Zak Murez, Nikhil Mohan, Sofía Dudas, Jeffrey Hawke, Vijay Badrinarayanan, Roberto Cipolla, Alex Kendall
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 temporal context (ablation) | FIERY · NuScenes future instance prediction | VPQ (Short): 40.3 -> 50.2 | Improved |
| add temporal context (ablation) | FIERY · NuScenes future instance prediction | IoU (Short): 51.7 -> 59.4 | Improved |
| add warping of past features to present frame (ablation) | FIERY · NuScenes future instance prediction | VPQ (Short): 41.7 -> 50.2 | Improved |
| add recursive unrolling of future states (ablation) | FIERY · NuScenes future instance prediction | VPQ (Short): 44.2 -> 50.2 | Improved |
| add future flow prediction (ablation) | FIERY · NuScenes future instance prediction | VPQ (Short): 44.6 -> 50.2 | Improved |
| add future flow prediction (ablation) | FIERY · NuScenes future instance prediction | IoU (Short): 58.0 -> 59.4 | Improved |
| add learned depth distribution lifting instead of uniform depth (ablation) | FIERY · NuScenes future instance prediction | VPQ (Short): 46.8 -> 50.2 | Improved |
| add learned depth distribution lifting instead of uniform depth (ablation) | FIERY · NuScenes future instance prediction | IoU (Short): 57.1 -> 59.4 | Hurt |
| add probabilistic modelling (present/future distributions) (ablation) | FIERY · NuScenes future instance prediction | VPQ (Short): 48.3 -> 50.2 | Improved |
| add probabilistic modelling (present/future distributions) (ablation) | FIERY · NuScenes future instance prediction | IoU (Long): 36.6 -> 36.7 | No clear effect |
| add past context (FIERY vs FIERY Static) | FIERY Static · NuScenes BEV segmentation (Setting 2) | IoU: 35.8 -> 38.2 | Improved |
| use FIERY's conditional-variational latent instead of Classical VAE prior | Classical VAE · NuScenes probabilistic modelling (GED) | Generalised Energy Distance (lower is better): 93.2 -> 90.5 | Improved |
Closest papers by meaning
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.89)
- Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting (NeurIPS 2021 · similarity 0.89)
- StretchBEV: Stretching Future Instance Prediction Spatially and Temporally (ECCV 2022 · similarity 0.88)
- FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving (NeurIPS 2025 · similarity 0.87)
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · similarity 0.86)
- DINO-Foresight: Looking into the Future with DINO (NeurIPS 2025 · similarity 0.85)
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.
