ZipMVS compresses cost volumes for memory-efficient multi-view stereo reconstruction
zhenjun_zhao · x · 2026-09-05
ZipMVS is a multi-view stereo method designed for efficient high-quality 3D reconstruction. A novel depth-hypothesis strategy — adjusting depth sampling ranges and deriving hypotheses from spatial cues — substantially compresses the cost volume, greatly reducing GPU memory while preserving accuracy. On DTU and Tanks and Temples it matches other efficiency-oriented MVS methods, balancing quality and memory. Code is available.
More from Research
- Metaⁿ and Recuris: filling the missing pieces in recursive self-improvement — TheTuringPost · 2026-09-05
- LLMs exploiting Lean bugs is a short-term problem, author argues — avt_im · 2026-09-05
- FinFIRST benchmark tests agents on real financial research: source discovery, evidence selection and math — alifcoder · 2026-09-05
- Valeo.ai Brings 5 ECCV 2026 Papers on Driving Video Prediction and LVLM Safety — abursuc · 2026-09-05
- Sakana AI's Percept-Lens: a simple rule on frozen vision features detects AI images — SakanaAILabs · 2026-09-05
- VI3 anchors pretrained 3D foundation models to metric scale using only IMU readings — zhenjun_zhao · 2026-09-05