// ----------------------------------------------------------------------------
// -                        Open3D: www.open3d.org                            -
// ----------------------------------------------------------------------------
// Copyright (c) 2018-2023 www.open3d.org
// SPDX-License-Identifier: MIT
// ----------------------------------------------------------------------------

#include "tensorflow/core/framework/op.h"
#include "tensorflow/core/framework/op_kernel.h"
#include "tensorflow/core/framework/shape_inference.h"
#include "tensorflow/core/lib/core/errors.h"

using namespace tensorflow;

REGISTER_OP("Open3DVoxelPooling")
        .Attr("TReal: {float, double}")  // type for the point positions
        .Attr("TFeat: {float, double, int32, int64}")  // type for the features
        .Attr("position_fn: {'average', 'nearest_neighbor', 'center'} = "
              "'average'")
        .Attr("feature_fn: {'average', 'nearest_neighbor', 'max'} = 'average'")
        .Attr("debug: bool = false")
        .Input("positions: TReal")
        .Input("features: TFeat")
        .Input("voxel_size: TReal")
        .Output("pooled_positions: TReal")
        .Output("pooled_features: TFeat")
        .SetShapeFn([](::tensorflow::shape_inference::InferenceContext* c) {
            using namespace ::tensorflow::shape_inference;
            ShapeHandle positions_shape, voxel_size_shape, features_shape,
                    pooled_positions_shape, pooled_features_shape;

            TF_RETURN_IF_ERROR(c->WithRank(c->input(0), 2, &positions_shape));
            TF_RETURN_IF_ERROR(c->WithRank(c->input(1), 2, &features_shape));
            TF_RETURN_IF_ERROR(c->WithRank(c->input(2), 0, &voxel_size_shape));

            // we don't know the number of output points
            pooled_positions_shape =
                    c->MakeShape({c->UnknownDim(), c->MakeDim(3)});
            c->set_output(0, pooled_positions_shape);

            DimensionHandle channel_dim = c->UnknownDim();
            if (c->RankKnown(features_shape)) {
                channel_dim = c->Dim(features_shape, -1);
            }
            pooled_features_shape =
                    c->MakeShape({c->UnknownDim(), channel_dim});
            c->set_output(1, pooled_features_shape);

            // check if we have a [N,3] tensor for the positions
            if (c->RankKnown(positions_shape)) {
                DimensionHandle d;
                TF_RETURN_IF_ERROR(
                        c->WithValue(c->Dim(positions_shape, -1), 3, &d));
            }

            return Status::OK();
        })
        .Doc(R"doc(
Spatial pooling for point clouds by combining points that fall into the same voxel bin.

The voxel grid used for pooling is always aligned to the origin (0,0,0) to
simplify building voxel grid hierarchies. The order of the returned voxels is
not defined as can be seen in the following example::

  import open3d.ml.tf as ml3d

  positions = [
      [0.1,0.1,0.1],
      [0.5,0.5,0.5],
      [1.7,1.7,1.7],
      [1.8,1.8,1.8],
      [0.3,2.4,1.4]]

  features = [[1.0,2.0],
              [1.1,2.3],
              [4.2,0.1],
              [1.3,3.4],
              [2.3,1.9]]

  ml3d.ops.voxel_pooling(positions, features, 1.0,
                         position_fn='center', feature_fn='max')

  # or with pytorch
  import torch
  import open3d.ml.torch as ml3d

  positions = torch.Tensor([
      [0.1,0.1,0.1],
      [0.5,0.5,0.5],
      [1.7,1.7,1.7],
      [1.8,1.8,1.8],
      [0.3,2.4,1.4]])

  features = torch.Tensor([
              [1.0,2.0],
              [1.1,2.3],
              [4.2,0.1],
              [1.3,3.4],
              [2.3,1.9]])

  ml3d.ops.voxel_pooling(positions, features, 1.0,
                         position_fn='center', feature_fn='max')

  # returns the voxel centers  [[0.5, 2.5, 1.5],
  #                             [1.5, 1.5, 1.5],
  #                             [0.5, 0.5, 0.5]]
  # and the max pooled features for each voxel [[2.3, 1.9],
  #                                             [4.2, 3.4],
  #                                             [1.1, 2.3]]

position_fn: Defines how the new point positions will be computed.
  The options are
    * "average" computes the center of gravity for the points within one voxel.
    * "nearest_neighbor" selects the point closest to the voxel center.
    * "center" uses the voxel center for the position of the generated point.

feature_fn: Defines how the pooled features will be computed.
  The options are
    * "average" computes the average feature vector.
    * "nearest_neighbor" selects the feature vector of the point closest to the voxel center.
    * "max" uses the maximum feature among all points within the voxel.

debug: If true additional checks for debugging will be enabled.

positions: The point positions with shape [N,3] with N as the number of points.

features: The feature vector with shape [N,channels].

voxel_size: The voxel size.

pooled_positions: The output point positions with shape [M,3] and M <= N.

pooled_features: The output point features with shape [M,channels] and M <= N.

)doc");

REGISTER_OP("Open3DVoxelPoolingGrad")
        .Attr("TReal: {float, double}")  // type for the point positions
        .Attr("TFeat: {float, double, int32, int64}")  // type for the features
        .Attr("position_fn: {'average', 'nearest_neighbor', 'center'} = "
              "'average'")
        .Attr("feature_fn: {'average', 'nearest_neighbor', 'max'} = 'average'")
        .Input("positions: TReal")
        .Input("features: TFeat")
        .Input("voxel_size: TReal")
        .Input("pooled_positions: TReal")
        .Input("pooled_features_gradient: TFeat")
        .Output("features_backprop: TFeat")
        .SetShapeFn([](::tensorflow::shape_inference::InferenceContext* c) {
            using namespace ::tensorflow::shape_inference;
            ShapeHandle positions_shape, voxel_size_shape, features_shape,
                    pooled_positions_shape, pooled_features_gradient_shape;

            TF_RETURN_IF_ERROR(c->WithRank(c->input(0), 2, &positions_shape));
            TF_RETURN_IF_ERROR(c->WithRank(c->input(1), 2, &features_shape));
            TF_RETURN_IF_ERROR(c->WithRank(c->input(2), 0, &voxel_size_shape));
            TF_RETURN_IF_ERROR(
                    c->WithRank(c->input(3), 2, &pooled_positions_shape));
            TF_RETURN_IF_ERROR(c->WithRank(c->input(4), 2,
                                           &pooled_features_gradient_shape));

            c->set_output(0, features_shape);

            DimensionHandle channel_dim = c->UnknownDim();
            if (c->RankKnown(features_shape)) {
                channel_dim = c->Dim(features_shape, -1);
            }
            if (c->RankKnown(pooled_features_gradient_shape)) {
                TF_RETURN_IF_ERROR(c->Merge(
                        channel_dim, c->Dim(pooled_features_gradient_shape, -1),
                        &channel_dim));
            }

            DimensionHandle first_dim = c->UnknownDim();
            if (c->RankKnown(features_shape)) {
                first_dim = c->Dim(features_shape, 0);
            }
            if (c->RankKnown(positions_shape)) {
                TF_RETURN_IF_ERROR(c->Merge(
                        first_dim, c->Dim(positions_shape, 0), &first_dim));
            }

            DimensionHandle first_dim_pooled = c->UnknownDim();
            if (c->RankKnown(pooled_features_gradient_shape)) {
                first_dim_pooled = c->Dim(pooled_features_gradient_shape, 0);
            }
            if (c->RankKnown(pooled_positions_shape)) {
                TF_RETURN_IF_ERROR(c->Merge(first_dim_pooled,
                                            c->Dim(pooled_positions_shape, 0),
                                            &first_dim_pooled));
            }

            // check if we have a [N,3] tensor for the positions
            if (c->RankKnown(positions_shape)) {
                DimensionHandle d;
                TF_RETURN_IF_ERROR(
                        c->WithValue(c->Dim(positions_shape, -1), 3, &d));
            }
            if (c->RankKnown(pooled_positions_shape)) {
                DimensionHandle d;
                TF_RETURN_IF_ERROR(c->WithValue(
                        c->Dim(pooled_positions_shape, -1), 3, &d));
            }

            return Status::OK();
        })
        .Doc(R"doc(
Gradient for features in VoxelPooling. For internal use only.
)doc");