Disparity Has a Sign: Stereo Matching Beyond the Zero-Disparity Plane

• URL: http://arxiv.org/abs/2609.06809v1
• Date: Sun, 06 Sep 2026 19:58:35 GMT
• Title: Disparity Has a Sign: Stereo Matching Beyond the Zero-Disparity Plane
• Abstract summary: We show how a stereoscopic content contains objects behind the zero-disparity plane corresponding to negative disparities.<n>We propose textitZDPShift, a benchmark of $21,495$ stereo pairs from seven cinematographer-authored open movies.<n>On identical scene content, Foundation goes from $2.24$ EPE to $75.33$px, with every backbone leaving roughly half of all pixels exceeding a three-pixel disparity error.
• Score: 52.210414985125816
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Modern stereo matching models fail when disparity crosses zero, with end-point error (EPE) rising by 4.6-37$\times$. Yet stereoscopic content, from cinema 3D to VR, routinely contains objects behind the zero-disparity plane (ZDP), corresponding to negative disparities. The blind spot cascades through datasets, architectures, and evaluation protocols, all of which inherit the non-negative geometry. Rectified parallel cameras place ZDP at infinity, so every finite depth yields $d=fB/z \ge 0$ by construction, and nothing within the standard pipeline can violate, or even measure, a negative disparity. To measure it, we propose \textit{ZDPShift}, a benchmark of $21{,}495$ stereo pairs from seven cinematographer-authored open movies, each frame rendered at five zero-disparity-plane positions with dense signed ground truth. Six state-of-the-art image and video stereo matching models collapse once the plane moves. On identical scene content, FoundationStereo goes from $2.24$ px EPE to $75.33$px, with every backbone leaving roughly half of all pixels exceeding a three-pixel disparity error. What is missing, however, is not the underlying matching capability. % The capability itself, however, is already present. Training on supervision synthesized from SceneFlow, which adds no new data or parameters, keeps the error flat across the signed range. Training only the decoder, with the pretrained matching features frozen, performs comparably across all six backbones, with EPE jittering within $0.2$px. Thus, the pretrained features already extend to the negative regime they were never trained on, and only the output convention discarded it. Meanwhile, positive-regime accuracy on KITTI, Middlebury, ETH3D, and Sintel is largely preserved.

Related papers

• CoVeR: Coverage-Based Token Pruning for Multi-View 3D Reasoning in VLMs [58.22241532542095]
  Representing a 3D scene as multi-view images allows 2D VLMs to reason in 3D by reusing priors from pre-training, sidestepping the scarcity of annotated 3D data.<n>Existing visual token pruners fall into two families, each limited in the 3D multi-view setting.<n>We show that spatial coverage is associated with 3D reasoning performance and introduce CoVeR, a deterministic, training-free selector that uses only token coordinates, with no learned signals.
  arXiv  Detail & Related papers  (2026-09-08T07:16:37Z)
• Elastic Token Compression for Pixel-Space Diffusion Transformers [64.18588817660013]
  Region Token Interface (Method) adapts a diffusion model to these tokens, with the region count drawn at random during fine-tuning.<n>Method leads prior reduction methods at matched budgets, matches dense quality at $2.0times$ the speed, and stays close at $2.6times$.
  arXiv  Detail & Related papers  (2026-08-29T14:06:27Z)
• Flex-$π$: A Multi-Stream World-Action Model with Compute Flexibility [25.477428304143235]
  We find a surprising free lunch: the same frozen video-generation VAE that encodes RGB also encodes 3D pointmaps almost losslessly, with no pointmap-specific training at all.<n>This lets us supervise Flex-$$, a 6B- parameter WAM, on 3D geometry and object-centric DINO semantics alongside RGB.<n>Every visual signal is projected into this shared latent space and denoised jointly with actions inside a Mixture-of-Transformers backbone.
  arXiv  Detail & Related papers  (2026-08-11T12:33:51Z)
• What CLIP Knows but Cannot Say: Recovering Negation from Frozen Intermediate Features [3.9492850032415876]
  Contrastive vision-language models such as CLIP map semantically opposite phrases to nearly identical embeddings.<n>We show that middle layers build compositional syntax, but the final layers collapse this structure as visual alignment rises.<n>We propose PeakPatch, a lightweight post-hoc correction system that intercepts the encoder at its compositional peak.
  arXiv  Detail & Related papers  (2026-07-25T16:15:50Z)
• More Structure, Not More Capacity: Object-Centric Representations for Visuomotor Imitation Learning [54.65906330923846]
  We show that object-centric slot representations are a structured alternative to pre-trained vision models.<n>A dense patch grid with 16x the tokens performs no better than the global feature.<n>An explicit 2D spatial goal and native-resolution rendering raises the full system to 68.7$pm$4.2%, just below a privileged 3D-oracle upper bound.
  arXiv  Detail & Related papers  (2026-07-10T10:35:24Z)
• RayTun3R: Online Camera Adaptation in 3D Foundation Models [53.97305597591864]
  Recent 3D foundation models provide strong feed-forward depth and pose estimates on pinhole imagery, but degrade sharply under fisheye camera geometry.<n>We show that this failure is partly caused by a pinhole camera bias in the positional encodings of pretrained 3D foundation models.<n>We propose RayTun3R, a lightweight camera adaptation approach.
  arXiv  Detail & Related papers  (2026-07-02T18:55:35Z)
• Cross-View Variance Correlation in Path-Traced Stereo:A Hidden Shortcut in Synthetic Training Data [0.0]
  Path-traced synthetic stereo data underlie a large fraction of disparity-estimation training pipelines.<n>We show that the underlying emphvariance fields are highly correlated once aligned by the ground-truth disparity warp.<n>This signal is unique to MC-rendered data and constitutes a candidate sim-to-real shortcut.
  arXiv  Detail & Related papers  (2026-06-24T07:13:00Z)
• Diffusion Transformer World-Action Model for AV Scene Prediction [0.0]
  Action-conditioned world models let an autonomous vehicle predict future camera scenes from its own planned controls.<n>But at compact, trainable scale the futures are ambiguous and the field's standard distortion metrics actively mislead.<n>We confront this with a compact latent world model that predicts future scene latents a frozen decoder renders to $256 times 256$ frames up to 8 seconds ahead.
  arXiv  Detail & Related papers  (2026-06-11T07:24:12Z)
• ZipSplat: Fewer Gaussians, Better Splats [89.08529100444004]
  We propose ZipSplat, a token-based feed-forward model that decouples Gaussian placement from the pixel grid.<n>Cross- and self-attention refine these tokens, and a single trained model spans the quality-efficiency curve without retraining.<n>ZipSplat operates without ground-truth poses or intrinsics, yet sets a new state of the art on DL3DV and RealEstate10K with $sim6times$ fewer Gaussians than pixel-aligned methods.
  arXiv  Detail & Related papers  (2026-06-03T17:04:30Z)
• Exact equivariance, kept through training, buys zero-shot generalisation across the symmetry group [6.230579198456525]
  A latent world model built from an equivariant encoder $E$ and an equivariant predictor $f$ inherits a provable symmetry of its training loss.<n>We verify this end-to-end at laptop scale (CPU/MPS, fully seeded)
  arXiv  Detail & Related papers  (2026-06-02T01:20:24Z)
• Self-Ensembling Gaussian Splatting for Few-Shot Novel View Synthesis [55.561961365113554]
  3D Gaussian Splatting (3DGS) has demonstrated remarkable effectiveness in novel view synthesis (NVS)<n>In this paper, we introduce Self-Ensembling Gaussian Splatting (SE-GS)<n>We achieve self-ensembling by incorporating an uncertainty-aware perturbation strategy during training.<n> Experimental results on the LLFF, Mip-NeRF360, DTU, and MVImgNet datasets demonstrate that our approach enhances NVS quality under few-shot training conditions.
  arXiv  Detail & Related papers  (2024-10-31T18:43:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.