Hartley & Zisserman, Chapter 8

More Single View Geometry

What a single image tells you about the 3D world. Vanishing points, vanishing lines, the image of the absolute conic, and camera calibration from a single photograph.

Prerequisites: Chapter 6 (Camera Models) + Chapter 7 (Camera Computation).
10
Chapters
5+
Simulations

Chapter 0: Why Single View Geometry?

You might think a single photograph is "flat" — that all 3D information is lost. But that is not true. A single image encodes a surprising amount of 3D structure, provided you know what to look for.

Parallel lines in the world converge to a vanishing point in the image. A set of parallel planes shares a vanishing line. These structures let you measure angles, distance ratios, and even recover the camera's internal calibration — all from one photograph.

The key insight: The camera projects points at infinity (directions in 3D) to finite image points (vanishing points). These vanishing points encode the geometry of the scene. From vanishing points alone, you can compute the focal length, principal point, and relative orientations of scene planes.
Vanishing Points

Parallel lines in 3D converge to a vanishing point in the image. The vanishing line of the ground plane passes through the vanishing points of all horizontal directions.

What is a vanishing point?

Chapter 1: Action of P on Geometric Entities

The camera matrix P acts not only on points, but on planes, lines, and conics. These actions form the vocabulary for single-view geometry.

EntityAction of PResult
3D point Xx = PXImage point
3D plane ππ = PTl (back-projection)The plane through camera centre and image line l
3D line (as join of two points)Image of two endpointsImage line
3D conic Q on a planeC = (Pπ)−T Q (Pπ)−1Image conic
Back-projection: A line l in the image back-projects to a plane PTl in 3-space. This plane contains the camera centre and all 3D points that could project to any point on l. A point x in the image back-projects to a ray through the camera centre.

The key to single-view geometry is exploiting these mappings. Vanishing points arise from projecting points at infinity (directions), and vanishing lines from projecting the line at infinity on a plane.

An image line l back-projects to what 3D entity?

Chapter 2: Vanishing Points

A vanishing point is the image of a point at infinity — that is, the image of a direction in 3D. If the direction vector is d, then the vanishing point is v = Md, where M is the left 3×3 submatrix of P.

Conversely, given a vanishing point v, the 3D direction is d = M−1v. This tells us the actual 3D direction of any set of parallel lines that converge to v.

Vanishing points from images: To find a vanishing point, detect two or more lines in the image that correspond to parallel lines in the world (e.g., edges of a road). Intersect these image lines. The intersection is the vanishing point for that direction.

An important property: the camera centre, the vanishing point, and any line in the corresponding 3D direction are coplanar. The ray from the camera centre through the vanishing point is parallel to the 3D lines.

Given a vanishing point v and the camera matrix M (left 3x3 of P), what is the 3D direction of the parallel lines?

Chapter 3: Vanishing Lines

A vanishing line is the image of the line at infinity on a plane. All vanishing points of directions lying in that plane fall on the vanishing line.

The vanishing line of a plane π is computed as l = (KKT)−1 v when you know the plane's normal direction, or more directly: if you have two vanishing points v1, v2 for two independent directions in the plane, then l = v1 × v2.

The horizon is a vanishing line. The horizon in a landscape photograph is the vanishing line of the ground plane. Every set of horizontal parallel lines (roads, railways, building edges) converges to a point on the horizon. The vertical vanishing point lies off the horizon (usually above or below the image).

The vanishing line encodes the orientation of the plane relative to the camera. From the vanishing line and one vanishing point perpendicular to the plane, you can determine the complete camera orientation.

How can you compute the vanishing line of a plane from the image?

Chapter 4: The Image of the Absolute Conic ω

The absolute conic Ω is a conic on the plane at infinity in 3-space. It is invariant under all Euclidean transformations (rotations and translations). Its image through a camera P = K[R|t] is:

ω = (KKT)−1

This is the image of the absolute conic (IAC). It depends only on K, not on R or t. So ω tells you the camera's intrinsics regardless of its pose.

Why ω matters: The angle θ between two rays through the camera centre with vanishing points v1 and v2 is:
cos θ = (v1T ω v2) / √(v1T ω v1 · v2T ω v2)

If you know ω, you can measure angles in 3D from a single image!

ω is a 3×3 symmetric matrix, defined up to scale, so it has 5 DOF — matching the 5 intrinsic parameters of K. To determine ω, we need 5 independent constraints.

The image of the absolute conic ω depends on which part of the camera matrix?

Chapter 5: Angles from ω

The power of ω is measuring angles. Two vanishing points v1, v2 correspond to orthogonal directions in 3D if and only if:

v1T ω v2 = 0

This is a linear constraint on ω. Every pair of orthogonal vanishing points gives one equation. Scenes are full of orthogonal structures (walls meet floors at right angles, window edges are perpendicular), providing rich constraints.

A vanishing point v and vanishing line l of a plane are orthogonal if l = ωv. This means the direction v is perpendicular to the plane. This is the relationship between the vertical vanishing point and the horizon line (the vanishing line of the ground plane).
Source of constraintType# constraints on ω
Orthogonal vanishing points v1, v2Linear1
Vanishing point v + vanishing line l at right anglesLinear2
Known metric plane (homography H)Linear2
Zero skew constraintLinear1
Square pixels constraintLinear2
Two vanishing points correspond to perpendicular 3D directions if and only if:

Chapter 6: Calibration from ω

Given enough constraints on ω, we can solve for it and then extract K via Cholesky factorization of ω−1 = KKT.

All the constraint types from the previous chapter are linear in the entries of ω. So the system Aw = 0 can be solved by SVD. With 5 or more constraints, ω is determined.

Example — calibration from 3 orthogonal vanishing points:
Assume zero skew and square pixels (2 constraints). Three pairs of orthogonal vanishing points give 3 more constraints. Total: 5. Enough to determine ω and hence K. The principal point turns out to be the orthocentre of the triangle formed by the three vanishing points.

If you also know the principal point is at the image centre and pixels are square, then only one unknown remains (the focal length f). A single pair of orthogonal vanishing points suffices to determine f.

Degeneracies arise when the constraints are not independent. For example, three metric planes that are all parallel provide only 2 independent constraints instead of 6.

With zero-skew and square-pixel assumptions, how many additional scene constraints are needed to determine ω?

Chapter 7: Affine Measurements

Single-view geometry enables real-world measurements from photographs. The most powerful technique measures length ratios of parallel line segments.

Given the vanishing line of the ground plane l, the vertical vanishing point v, and two vertical line segments in the image with known base positions:

StepAction
1Transfer the length of one segment onto the other using a line through the vanishing point of the base-to-base direction
2Use the four collinear image points (base, transferred top, actual top, vanishing point) to compute the cross-ratio and hence the length ratio
Measuring a person's height from one photo: If you know the horizon line and the vertical vanishing point, plus one object of known height (e.g., a door frame at 2.1m), you can measure anyone's height to within ~2cm accuracy — provided their head and feet are visible.

No knowledge of the camera calibration K is needed for these ratio measurements. The vanishing points and vanishing line carry all the necessary information.

To measure length ratios of parallel segments from a single image, what do you need?

Chapter 8: Plane Rectification

Given the vanishing line of a plane, we can rectify the image of that plane — warp the image so that the plane appears fronto-parallel (as if photographed straight-on).

The rectifying homography is determined by:

H = K Rcorrect R−1 K−1

If K is known, the vanishing line determines the plane orientation, and we can compute Rcorrect to rotate the plane to fronto-parallel. Even without K, if we know the aspect ratio of a rectangle on the plane, we can compute the full rectification.

Applications: Rectification is used to remove perspective distortion from building facades, paintings, documents, and road signs. If you know the camera has square pixels and can identify the vanishing line, you only need the focal length — which can be computed from a single pair of orthogonal vanishing points.
What does plane rectification achieve?

Chapter 9: Connections

Single-view geometry is the bridge between camera models and multi-view geometry. It shows how much information a camera matrix P encodes about the 3D world.

ConceptWhere it leads
Vanishing pointsCh 9: Epipoles are vanishing points of the translation direction
IAC ωCh 19: Auto-calibration recovers ω from multiple views without a calibration target
Plane rectificationCh 11: Image rectification makes epipolar lines horizontal for stereo matching
Back-projectionCh 12: Triangulation intersects back-projected rays from two views
"The image of the absolute conic ω encodes the camera's internal geometry. Once ω is known, angles between any two rays can be measured."
— Hartley & Zisserman, Chapter 8
What geometric entity encodes all 5 intrinsic camera parameters and is invariant to camera pose?
← Chapter 7 Chapter 9: Epipolar Geometry →