AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

Efficient NeRF Optimization - Not All Samples Remain Equally Hard

Juuso Korhonen, Goutham Rangu, Hamed Rezazadegan Tavakoli, Juho Kannala

ECCV 2024 · 3D vision › NeRF and radiance fields · approach: Dynamic and editable NeRF · 3 citations (OpenAlex)

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 changedWhereMeasuredEffect
mine hard samples with an inference-mode first forward pass and backpropagate only b hard samplesInstant-NGP (PyTorch) · Mip-NeRF-360validation PSNR per training time: ~1 dB gain per same training time; ~2x faster to a given PSNRImproved
build the computational graph only for hard samplesInstant-NGP (PyTorch) · Mip-NeRF-360network memory: ~1 GB -> ~570 MBImproved
train on hard sample minibatch (b/B about 10-15%)Instant-NGP (PyTorch) · Mip-NeRF-360iteration time: down to ~53-54% of baselineImproved
hard sample mining versus baseline per iterationInstant-NGP (PyTorch) · Mip-NeRF-360training loss per iterationNo clear effect
apply hard sample mining on scenes with uniformly hard content (room, garden, stump)Instant-NGP (PyTorch) · Mip-NeRF-360PSNR gain: stump 8 min 23.4 vs 23.4; bonsai 29.7 vs 28.4Mixed
train longer (20 min) with hard sample miningInstant-NGP (PyTorch) · Mip-NeRF-360 stumpvalidation PSNR: 23.28 vs 23.46 baselineHurt
apply hard sample mining to Nerfacto with proposal samplingNerfacto · Mip-NeRF-360PSNR per training time and memory: memory 300 MB -> 160 MB; PSNR gains smallerMixed

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy