Research map / Federated learning and privacy / Privacy attacks and secure learning
Private Multiparty Perception for Navigation
Hui Lu, Mia Chiquier, Carl Vondrick
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 multiview camera inputs and perform inference with two-party MPC | CipherNav navigation model · Obstacle World test, Overall | accuracy: 74.0% -> 96.6% | Improved |
| Add multiview camera inputs and perform inference with five-party MPC | CipherNav navigation model · Obstacle World test, Overall | accuracy: 74.0% -> 96.9% | Improved |
| Replace plaintext multiview inference with two-party MPC | Navigation with camera views · Obstacle World test, Overall | accuracy: 97.1% -> 96.6% | Hurt |
| Replace plaintext multiview inference with five-party MPC | Navigation with camera views · Obstacle World test, Overall | accuracy: 97.1% -> 96.9% | Hurt |
| Use a deterministic rather than random starting view direction | First Person baseline · Obstacle World test, Overall | accuracy: 72.0% -> 88.8% | Improved |
| Add external camera views to map and agent inputs | Navigation model · Obstacle World successful paths, Overall | path efficiency: 95.5 % -> 99.5 % | Improved |
| Use two-party MPC instead of plaintext for multiview inference | Navigation with camera views · Obstacle World successful paths, Overall | path efficiency: 99.4 % -> 99.5 % | Mixed |
| Use five-party MPC instead of plaintext for multiview inference | Navigation with camera views · Obstacle World successful paths, Overall | path efficiency: 99.4 % -> 99.2 % | Hurt |
| Perform inference with two-party MPC rather than plaintext | CipherNav action prediction network · Obstacle World runtime, batch size 100, GPU | runtime (seconds): 0.00028 -> 0.15 | Hurt |
| Perform inference with five-party MPC rather than plaintext | CipherNav action prediction network · Obstacle World runtime, batch size 100, GPU | runtime (seconds): 0.00028 -> 0.42 | Hurt |
| Use encrypted rather than plaintext camera features | Scene reconstruction classifier · Obstacle World scene reconstruction | accuracy: 93.6% -> 49.7% | Hurt |
Closest papers by meaning
- Instance-aware Exploration-Verification-Exploitation for Instance ImageGoal Navigation (CVPR 2024 · similarity 0.85)
- Personalized Instance-based Navigation Toward User-Specific Objects in Realistic Environments (NeurIPS 2024 · similarity 0.84)
- Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction (ICLR 2024 · similarity 0.84)
- Habitat Synthetic Scenes Dataset (HSSD-200): An Analysis of 3D Scene Scale and Realism Tradeoffs for ObjectGoal Navigation (CVPR 2024 · similarity 0.83)
- Privacy-Preserving V2X Collaborative Perception Integrating Unknown Collaborators (AAAI 2025 · similarity 0.83)
- Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator (NeurIPS 2026 · similarity 0.83)
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.
