[object Object]

[object Object][object Object]undefined[object Object]

• Description: Performs rasterization computation. It obtains the minimum depth and the corresponding face index of each pixel on the screen based on the given points and faces in the 3D space, and calculates the perspective-corrected barycentric coordinates of the face.

• Formulas: $findices$ records the face index corresponding to the minimum depth of each pixel, and $barycentric$ records the perspective-corrected barycentric coordinates of each vertex relative to the face recorded in $findices$. The Z-buffer used during the calculation records the minimum depth $z_{\min}(x, y)$ of each pixel $(x, y)$ and the index $\text{face\_idx}(x, y)$ of the triangle face corresponding to that depth.

  The computation process is as follows: For each triangle face $f$ in the space:

  1. Convert the coordinates of the three vertices $v_0$, $v_1$, and $v_2$ of $f$ to the screen coordinates $v_{s0}$, $v_{s1}$, and $v_{s2}$.

  2. Calculate the rectangular area that encloses $f$ based on $v_{s0}$, $v_{s1}$, and $v_{s2}$.

  3. For each pixel $v_i = (x_i, y_i)$ in the rectangle, perform the following operations:

     a. Calculate the center coordinates $v_c$ of the pixel. b. Calculate the barycentric coordinates $(\alpha, \beta, \gamma)$ of $v_c$ relative to the triangle $f$. c. Determine whether $v_c$ is inside the triangle based on $(\alpha, \beta, \gamma)$. If $v_c$ is not inside the triangle, process the next pixel in the rectangle. Otherwise, go to the next step. d. Use $(\alpha, \beta, \gamma)$ and $v_{s0}$, $v_{s1}$, $v_{s2}$ to obtain the depth value of the current pixel. e. If the depth prior is enabled, update the Z-buffer. Otherwise, go to the next step.

     • Calculate the depth threshold depth_thres using the depth prior map.
     • If depth < depth_thres, process the next pixel in the rectangle. Otherwise, go to the next step.

     f. Z-buffer update:

     • If $depth < z_{\min}(x_i, y_i)$:
     $$\quad z_{\min}(x_i, y_i) \gets \text{depth} \\ \quad \text{face\_idx}(x_i, y_i) \gets f$$
     • If $depth = z_{\min}(x_i, y_i)$:
     $$\quad \text{face\_idx}(x_i, y_i) \gets \min(\text{face\_idx}(x_i, y_i),\ f)$$

  After all triangle patches in the space are processed according to the preceding steps, for each pixel $v_i = (x_i, y_i)$ on the screen with the size of $height \times width$:

  1. Obtain the patch index $f_{idx}$ corresponding to $v_i$ in the Z-buffer, and set $findices (x_i, y_i) \gets f_{idx}$.
  2. Convert the coordinates of the three vertices $v_0$, $v_1$, and $v_2$ of $f$ to the screen coordinates $v_{s0}$, $v_{s1}$, and $v_{s2}$.
  3. Calculate the coordinates of the center point $v_c$ of $v_i$.
  4. Calculate the barycentric coordinates $(\alpha, \beta, \gamma)$ of $v_c$ relative to the triangle $f$.
  5. Use $(\alpha, \beta, \gamma)$ to calculate the perspective-corrected interpolation $(\tilde{\alpha}, \tilde{\beta}, \tilde{\gamma})$.
  6. $barycentric(x_i, y_i) \gets (\tilde{\alpha}, \tilde{\beta}, \tilde{\gamma})$

  The following are the involved calculation methods:

  • Convert the vertex $v = (x, y, z, w)$ to the screen coordinates $v_s = (x_s, y_s, z_s)$.

    $$x_s = (x / w * 0.5 + 0.5) * (width - 1) + 0.5\\ y_s = (0.5 + 0.5 * y / w) * (height - 1) + 0.5\\ z_s = z / w * 0.49999 + 0.5$$

  • Barycentric coordinates $(\alpha, \beta, \gamma)$ of point $v$ relative to the triangle $(v_0, v_1, v_2)$

    1. Calculate the directed areas $area$, $beta\_tri$, and $gamma\_tri$ of triangles $(v_0, v_1, v_2)$, $(v_0, v, v_2)$, and $(v_0, v_1, v)$, respectively.
    2. If $area$ is 0, then $\alpha = \beta = \gamma = -1$. Otherwise,
    $$\beta = beta\_tri / area\\ \gamma = gamma\_tri / area\\ \alpha = 1 - \beta - \gamma$$

  • The directed area of the triangle formed by vertices $v_0 = (x_0, y_0, z_0)$, $v_1 = (x_1, y_1, z_1)$, and $v_2 = (x_2, y_2, z_2)$.

    $$

  are a = (x_2 - x_0) * (y_1 - y_0) - (x_1 - x_0) * (y_2 - y_0) $$

  • Calculate the depth $depth$ of pixel $v = (x, y)$ based on the barycentric coordinates $(\alpha, \beta, \gamma)$ and the screen coordinates of the triangle $v_0 = (x_0, y_0, z_0)$, $v_1 = (x_1, y_1, z_1)$, and $v_2 = (x_2, y_2, z_2)$.

    $$depth = \alpha * z_0 + \beta * z_1 + \gamma * z_2$$

  • Calculate the depth threshold $depth\_thres$ of point $v = (x, y)$ based on the depth map $d$ and occlusion truncation $occlusion\_truncation$.

    $$depth\_thres = d(x, y) * 0.49999 + 0.5 + occlusion\_truncation$$

  • Determine whether the vertex is inside the triangle based on the barycentric coordinates $(\alpha, \beta, \gamma)$. If $\alpha >= 0$, $\beta >= 0$, and $\gamma >= 0$, the point is inside the triangle (including on the triangle edges). Otherwise, the point is not inside the triangle.

  • Calculate the perspective correction interpolation $(\lambda_0^{corrected}, \lambda_1^{corrected}, \lambda_2^{corrected})$ based on the barycentric coordinates $(\lambda_0, \lambda_1, \lambda_2)$ and the coordinates of the three vertices of the triangle $v_0 = (x_0, y_0, z_0, w_0)$, $v_1 = (x_1, y_1, z_1, w_1)$, and $v_2 = (x_2, y_2, z_2, w_2)$.

    $$\lambda_i^{corrected} = \frac{\lambda_i / w_i} { \sum (\lambda_j / w_j)}$$

[object Object]

Each operator has calls. First, [object Object] is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, [object Object] is called to perform computation.

[object Object]

• Parameters

  [object Object]

• Returns:

  aclnnStatus: return status code. For details, see .

  The first-phase API implements input parameter validation. The following error codes may be returned.

  [object Object]

[object Object]

• Parameters

  [object Object]

• Returns:

  [object Object]: status code. For details, see .

[object Object]

• Only the scenario where useDepthPrior is 0 is supported. The parameters dOptional, occlusionTruncation, and useDepthPrior do not take effect in actual computation.

• Deterministic computation:

  • The aclnnRasterizer is implemented in deterministic mode by default.

[object Object]

The following example is for reference only. For details, see .