matcher.py

# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
"""
Mostly copy-paste from DETR (https://github.com/facebookresearch/detr).
"""
import torch
from scipy.optimize import linear_sum_assignment
from torch import nn


class HungarianMatcher_Crowd(nn.Module):
    """This class computes an assignment between the targets and the predictions of the network

    For efficiency reasons, the targets don't include the no_object. Because of this, in general,
    there are more predictions than targets. In this case, we do a 1-to-1 matching of the best predictions,
    while the others are un-matched (and thus treated as non-objects).
    """

    def __init__(
        self,
        cost_class: float = 1,
        cost_point: float = 1,
        override_multiclass: bool = False,
        pointmatch: bool = False
    ):
        """Creates the matcher

        Params:
            cost_class: This is the relative weight of the foreground object
            cost_point: This is the relative weight of the L1 error of the points coordinates in the matching cost
        """
        super().__init__()
        self.cost_class = cost_class
        self.cost_point = cost_point
        assert cost_class != 0 or cost_point != 0, "all costs cant be 0"
        self.override_multiclass = override_multiclass
        self.pointmatch = pointmatch

    @torch.no_grad()
    def forward(self, outputs, targets):
        """Performs the matching

        Params:
            outputs: This is a dict that contains at least these entries:
                 "pred_logits": Tensor of dim [batch_size, num_queries, num_classes] with the classification logits
                 "points": Tensor of dim [batch_size, num_queries, 2] with the predicted point coordinates

            targets: This is a list of targets (len(targets) = batch_size), where each target is a dict containing:
                 "labels": Tensor of dim [num_target_points] (where num_target_points is the number of ground-truth
                           objects in the target) containing the class labels
                 "points": Tensor of dim [num_target_points, 2] containing the target point coordinates

        Returns:
            A list of size batch_size, containing tuples of (index_i, index_j) where:
                - index_i is the indices of the selected predictions (in order)
                - index_j is the indices of the corresponding selected targets (in order)
            For each batch element, it holds:
                len(index_i) = len(index_j) = min(num_queries, num_target_points)
        """
        bs, num_queries = outputs["pred_logits"].shape[:2]

        # We flatten to compute the cost matrices in a batch
        out_prob = (
            outputs["pred_logits"].flatten(0, 1).softmax(-1)
        )  # [batch_size * num_queries, num_classes]
        out_points = outputs["pred_points"].flatten(
            0, 1
        )  # [batch_size * num_queries, 2]
        # Also concat the target labels and points
        # tgt_ids = torch.cat([v["labels"] for v in targets])

        tgt_ids = torch.cat([v["labels"] for v in targets])
        tgt_points = torch.cat([v["point"] for v in targets])

        if self.override_multiclass:
            tgt_ids = torch.ones(tgt_ids.size()[0], dtype=torch.int)
        # Compute the classification cost. Contrary to the loss, we don't use the NLL,
        # but approximate it in 1 - proba[target class].
        # The 1 is a constant that doesn't change the matching, it can be ommitted.
        cost_class = -out_prob[:, tgt_ids]
        # Compute the L2 cost between point
        cost_point = torch.cdist(out_points, tgt_points, p=2)

        # Compute the GIoU cost between point
        # Final cost matrix
        if self.pointmatch:
            C = cost_point
        else:
            C = self.cost_point * cost_point + self.cost_class * cost_class
        
        # Reshape to back to [batch_size, num_queries, num_target_points]
        C = C.view(bs, num_queries, -1).cpu()
        # compute the matching
        sizes = [len(v["point"]) for v in targets]
        indices = [
            linear_sum_assignment(c[i]) for i, c in enumerate(C.split(sizes, -1))
        ]

        return [
            (
                torch.as_tensor(i, dtype=torch.int64),
                torch.as_tensor(j, dtype=torch.int64),
            )
            for i, j in indices
        ]


def build_matcher_crowd(args, override_multiclass: bool = False):
    return HungarianMatcher_Crowd(
        cost_class=args.set_cost_class,
        cost_point=args.set_cost_point,
        override_multiclass=override_multiclass,
        pointmatch=args.pointmatch,
    )