TriINR: Unsupervised Atmospheric Turbulence Mitigation with Triple Implicit Neural Representations
To address this limitation, we propose TriINR, an unsupervised turbulence mitigation method that recovers a sharp image from multiple degraded frames.
ProofPaper ↗
Key points
- Atmospheric turbulence degrades image quality in long-range imaging.
- Recent unsupervised turbulence mitigation methods have demonstrated the effectiveness of representing a sharp image as an implicit neural representation (INR).
- By modeling tilt and blur as spatio-temporal INRs, TriINR achieves frame-count-independent degradation representation and memory-efficient optimization.
- Experiments on synthesized turbulence data with three degradation strengths show that TriINR restores images more robustly across varying turbulence strengths than prior TM methods and achieves the highest accuracy among the evaluated unsupervised methods.
Sources (1)
- [1]TriINR: Unsupervised Atmospheric Turbulence Mitigation with Triple Implicit Neural RepresentationsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 16, 01:35 PM
To address this limitation, we propose TriINR, an unsupervised turbulence mitigation method that recovers a sharp image from multiple degraded frames.
Atmospheric turbulence degrades image quality in long-range imaging.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 16, 2026Look, Reason, Draw a Region: Recursive Refinement for Spatial Context Prediction
- Oct 16, 2026Persistent Planar Memory for Video World Models
- Oct 15, 2026REKI: Registering Knowledge in VLAs
- Oct 15, 2026FALCON: FPGA-Accelerated Latency-Certifiable Occupancy Networks for Camera-Based Autonomous Driving
- Oct 15, 2026NVIDIA Nemotron 3.5 Super VL Powers Specialized Agents with Multimodal Reasoning
- Oct 15, 2026Learning to Think and Explore in 3D World