Slice-Loc: Cross-View Localization & Beyond

• Slice-Loc is a multi-context framework that uses redundant image slicing and a-contrario statistical validation for precise cross-view 3-DoF pose estimation.
• In applied harmonic analysis, it leverages a local Fourier slice theorem with sparse wavelet frames to enable efficient local signal reconstruction.
• In knot theory, Slice-Loc examines slice disk classification by distinguishing non-equivalent disks through group-invariant analysis and local modifications.

Slice-Loc is a term appearing in several specialized domains. In computer vision, it designates a two-stage cross-view localization strategy based on redundant observations using image slicing and robust validation via a-contrario statistical modeling. In analysis, Slice-Loc refers to a local projection theorem leveraging wavelet bases. In knot theory, Slice-Loc is connected to the study of slice disks, especially their classification up to local modifications. This entry covers each of these contexts, with a primary focus on the cross-view localization methodology, as presented in recent literature.

1. Cross-View Localization via Redundant Slicing and A-Contrario Validation

Cross-view localization (CVL) seeks to align ground-level panoramic images with georeferenced aerial viewpoints to recover the 3-DoF camera pose—planar position and azimuth—especially for scenarios where GNSS is denied. Conventional CVL methods provide a single pose estimate per query, making it difficult to robustly assess localization validity, detect outliers, or guarantee reliability under ambiguous conditions. Slice-Loc introduces redundancy by partitioning the ground image into overlapping sub-images ("slices"), each producing an independent 3-DoF pose candidate, which are later aggregated and statistically validated for geometric coherence (Zhang et al., 7 Aug 2025).

Two-Stage Architecture

Slice-Loc consists of two principal stages:

• Stage 1: Slicing and 3-DoF Estimation The ground panorama $I_G$ is split into $n$ overlapping slices $G_i$ (typically $n=12$), each corresponding to a distinct horizontal field-of-view window. Each slice is independently projected to a pinhole view and processed by a feature-based deep network (CCVPE) to predict planar coordinates $(x_i, y_i)$ in the aerial map and a relative orientation $\phi_i$. Rotational priors are enforced by pre-aligning the aerial reference for each slice's field of view.
• Stage 2: A-Contrario Validation (OSA-CVL) All slice predictions $\mathcal S_G = \{\hat p_i\}$ are jointly assessed: candidate camera locations $p_c$ are hypothesized from slice pairs, and for each $p_c$, the geometric rigidity is measured by the angular deviation $\theta_i$ between predicted slice ray and expected geometry. Subsets of slices are scored using the Number of False Alarms (NFA), which quantifies the statistical likelihood of such geometric agreement under a uniform null; the globally minimal NFA determines the most reliable inlier set, and a threshold $n$0 is used to accept or reject the localization hypothesis.

2. Slicing Geometry, 3-DoF Prediction, and Ground-Truth Construction

The panoramic ground image of width $n$1 and height $n$2 is mapped from pixel to spherical coordinates by $n$3, $n$4. Slice centers are evenly distributed in azimuth, each covering $n$5 with common elevation. Each slice is rendered into a $n$6 pinhole projection. For pose estimation, a Siamese/transformer network (CCVPE) extracts features from $n$7 and the aerial reference $n$8, regressing both position and orientation. Training supervision uses geometry-based ground-truthing: pixelwise projections via depth maps, with slice ground-truth locations defined as the mean of valid reprojected points.

3. Geometric Rigidity Metric and Robust Inlier Selection

Given a hypothesized camera position $n$9, each slice's predicted scene point should lie along the direction of the slice's central ray from $G_i$0. The principal angular error is

$G_i$1

For a subset $G_i$2 of size $G_i$3, the best $G_i$4 minimizes total angular error. The set's rigidity is defined as the maximal slice error,

$G_i$5

This metric is used as the basis for inlier selection and meaningfulness estimation in the a-contrario framework.

4. A-Contrario Statistical Model and NFA Computation

The a-contrario model postulates a null hypothesis $G_i$6: if slice localizations are random with respect to visual cues, their predicted positions are uniformly distributed in an admissible search sector. For an allowed angular error $G_i$7, the cumulative distribution function $G_i$8 is empirically fitted, typically via piecewise linear approximations from simulation data.

For a subset $G_i$9 of $n=12$0 slices, the expected NFA is

$n=12$1

where combinatorial factors account for choice of zero-error reference, subset sampling, and candidate camera pose definition. A set is declared meaningful if $n=12$2. A fixed threshold $n=12$3 is recommended.

Notation	Definition	Role
$n=12$4	CDF of angular error under $n=12$5	Localization likelihood
$n=12$6	Expected number of false alarms (NFA)	Statistical test threshold
$n=12$7	Subset of slice pose predictions	Inlier set

5. Algorithmic Implementation and Quantitative Results

Algorithm Summary

• Step A: Slice the panorama and predict 3-DoF poses $n=12$8 using CCVPE.
• Step B: For all slice pairs, estimate candidate $n=12$9, sort slices by geometric error, and iterate over $(x_i, y_i)$0 to compute NFA for each inlier set. Retain the global minimum NFA and corresponding inlier subset.
• Step C: Accept the localization only if $(x_i, y_i)$1, then re-optimize using inliers.

Implementation Specifications

• Backbone: CCVPE deep feature network.
• Major parameters: $(x_i, y_i)$2 slices (30° intervals), slice $(x_i, y_i)$3 FoV, reference aerial tiles $(x_i, y_i)$4 px at 0.11 m/px.
• Training: AdamW, batch size 12, initial learning rate $(x_i, y_i)$5, 12 epochs.
• Hardware: Single GPU (e.g., NVIDIA RTX 4090).
• A-contrario fit: Fixed breakpoints $(x_i, y_i)$6; fit parameters $(x_i, y_i)$7.

Quantitative Evaluation

On the DReSS Cross-Area (±45° rotation noise):

Method	Mean loc. error (m)	Mean orient. error (°)
CCVPE (baseline)	4.47	3.42
Slice-Loc (pre-filter)	4.23	1.87
Slice-Loc (filtered)	1.86	1.24

Slice-Loc achieves a 58% reduction in localization error and a 64% reduction in orientation error after a-contrario filtering. Less than 3% of accepted cases exceed 10 m error. On VIGOR, Slice-Loc further cuts the mean error to 1.65 m, with >91% of results under 3 m (Zhang et al., 7 Aug 2025).

6. Mathematical Formulation in Local Fourier Slice Theory

In applied harmonic analysis, the “Slice-Loc” moniker also refers to a local Fourier slice theorem for signal projections in sparse wavelet frames (Lessig, 2018). A function $(x_i, y_i)$8 expanded in polar wavelet frames can be locally projected along arbitrary directions, using analytically computed “sliced” wavelets: $(x_i, y_i)$9 Here, $\phi_i$0 is a lower-dimensional wavelet obtained by slicing the angular part in frequency. This formalism supports efficient local/sparse reconstructions for tomography and generalizes the classical Fourier slice theorem.

7. Slice-Loc in Knot Theory: Equivalence of Slice Disks

In low-dimensional topology, Slice-Loc describes the classification of slice disks of a fixed knot $\phi_i$1 in $\phi_i$2, modulo local knotting operations. Disks are declared equivalent if they differ only by connected sums with 2-knots embedded in $\phi_i$3; two disks with different kernels in the map

$\phi_i$4

are not equivalent in this sense. Recent classification results show that for many constructions—twist-spun disks, generalized square knot disks, and certain fibered hyperbolic examples—there exist infinite families of pairwise non-equivalent slice disks distinguished by their group-theoretical invariants (Meier et al., 12 Mar 2025).

• "Cross-View Localization via Redundant Sliced Observations and A-Contrario Validation" (Zhang et al., 7 Aug 2025)
• "A Local Fourier Slice Theorem" (Lessig, 2018)
• "Slice disks modulo local knotting" (Meier et al., 12 Mar 2025)