AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Pip-Stereo: Progressive Iterations Pruner for Iterative Optimization based Stereo Matching

Jintu Zheng, Qizhe Liu, Huangxin Xu, Zhuojie Chen

CVPR 2026 · Architectures and efficiency › RNNs and normalization · approach: Dataset condensation and flow estimation · 0 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
add monocular depth prior transfer (MPT) (ablation)Selective-IGEV · SceneFlow testEPE: 0.44 -> 0.38 (32 iters)Improved
add supernet search for student block allocation (ablation)PipStereo MPT · SceneFlow testEPE: 0.40 -> 0.38Improved
add MPT when using a single inference iteration (ablation)Selective-IGEV · SceneFlow testEPE: 0.64 -> 0.56Improved
add PIP finetuning when pruning to 1 iteration (ablation)PipStereo-i32 · SceneFlow testEPE: 0.56 -> 0.45Improved
prune iterations from 32 to 16 with PIP (ablation)PipStereo · SceneFlow testEPE: 0.38 -> 0.38No clear effect
prune iterations further (16 to 1) with PIP (ablation)PipStereo · SceneFlow testEPE: 0.38 -> 0.45Hurt
add PIP to single-iteration FoundationStereo (ablation)FoundationStereo · ETH3DBad-1 (Noc): 0.94 -> 0.70Improved
add PIP to single-iteration RAFT-Stereo (ablation)RAFT-Stereo · SceneFlow testEPE: 3.65 -> 2.16Mixed
add FlashGRU to PIP-pruned model (8 iters) (ablation)PipStereo-p2 · SceneFlow latencylatency (s): 1.02 -> 0.76Improved
add FlashGRU to PIP-pruned RAFT-Stereo (4 iters) (ablation)Raft-p3 · SceneFlow latencylatency (s): 1.07 -> 0.69Improved
add FlashGRU to ConvGRU at 2K resolutionSelective-IGEV update block · RTX 4090 1280x2944RNN latency: 122.4ms -> 16.8ms (x7.28)Improved
add FlashGRU at low resolutionSelective-IGEV update block · RTX 4090 320x736RNN latency: 28.8ms -> 14.8ms (x1.94)Improved

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