| """ |
| Patch Projector |
| """ |
| import torch |
| import torch.nn as nn |
| import torch.nn.functional as F |
| import numpy as np |
| from models.render_utils import sample_ptsFeatures_from_featureMaps |
|
|
|
|
| class PatchProjector(): |
| def __init__(self, patch_size): |
| self.h_patch_size = patch_size |
| self.offsets = build_patch_offset(patch_size) |
|
|
| self.z_axis = torch.tensor([0, 0, 1]).float() |
|
|
| self.plane_dist_thresh = 0.001 |
|
|
| |
| def pixel_warp(self, pts, imgs, intrinsics, |
| w2cs, img_wh=None): |
| """ |
| |
| :param pts: [N_rays, n_samples, 3] |
| :param imgs: [N_views, 3, H, W] |
| :param intrinsics: [N_views, 4, 4] |
| :param c2ws: [N_views, 4, 4] |
| :param img_wh: |
| :return: |
| """ |
| if img_wh is None: |
| N_views, _, sizeH, sizeW = imgs.shape |
| img_wh = [sizeW, sizeH] |
|
|
| pts_color, valid_mask = sample_ptsFeatures_from_featureMaps( |
| pts, imgs, w2cs, intrinsics, img_wh, |
| proj_matrix=None, return_mask=True) |
|
|
| pts_color = pts_color.permute(2, 3, 0, 1) |
| valid_mask = valid_mask.permute(1, 2, 0) |
|
|
| return pts_color, valid_mask |
|
|
| def patch_warp(self, pts, uv, normals, src_imgs, |
| ref_intrinsic, src_intrinsics, |
| ref_c2w, src_c2ws, img_wh=None |
| ): |
| """ |
| |
| :param pts: [N_rays, n_samples, 3] |
| :param uv : [N_rays, 2] normalized in (-1, 1) |
| :param normals: [N_rays, n_samples, 3] The normal of pt in world space |
| :param src_imgs: [N_src, 3, h, w] |
| :param ref_intrinsic: [4,4] |
| :param src_intrinsics: [N_src, 4, 4] |
| :param ref_c2w: [4,4] |
| :param src_c2ws: [N_src, 4, 4] |
| :return: |
| """ |
| device = pts.device |
|
|
| N_rays, n_samples, _ = pts.shape |
| N_pts = N_rays * n_samples |
|
|
| N_src, _, sizeH, sizeW = src_imgs.shape |
|
|
| if img_wh is not None: |
| sizeW, sizeH = img_wh[0], img_wh[1] |
|
|
| |
| uv[:, 0] = (uv[:, 0] + 1) / 2. * (sizeW - 1) |
| uv[:, 1] = (uv[:, 1] + 1) / 2. * (sizeH - 1) |
|
|
| ref_intr = ref_intrinsic[:3, :3] |
| inv_ref_intr = torch.inverse(ref_intr) |
| src_intrs = src_intrinsics[:, :3, :3] |
| inv_src_intrs = torch.inverse(src_intrs) |
|
|
| ref_pose = ref_c2w |
| inv_ref_pose = torch.inverse(ref_pose) |
| src_poses = src_c2ws |
| inv_src_poses = torch.inverse(src_poses) |
|
|
| ref_cam_loc = ref_pose[:3, 3].unsqueeze(0) |
| sampled_dists = torch.norm(pts - ref_cam_loc, dim=-1) |
|
|
| relative_proj = inv_src_poses @ ref_pose |
| R_rel = relative_proj[:, :3, :3] |
| t_rel = relative_proj[:, :3, 3:] |
| R_ref = inv_ref_pose[:3, :3] |
| t_ref = inv_ref_pose[:3, 3:] |
|
|
| pts = pts.view(-1, 3) |
| normals = normals.view(-1, 3) |
|
|
| with torch.no_grad(): |
| rot_normals = R_ref @ normals.unsqueeze(-1) |
| points_in_ref = R_ref @ pts.unsqueeze( |
| -1) + t_ref |
| d1 = torch.sum(rot_normals * points_in_ref, dim=1).unsqueeze( |
| 1) |
|
|
| d2 = torch.sum(rot_normals.unsqueeze(1) * (-R_rel.transpose(1, 2) @ t_rel).unsqueeze(0), |
| dim=2) |
| valid_hom = (torch.abs(d1) > self.plane_dist_thresh) & ( |
| torch.abs(d1 - d2) > self.plane_dist_thresh) & ((d2 / d1) < 1) |
|
|
| d1 = d1.squeeze() |
| sign = torch.sign(d1) |
| sign[sign == 0] = 1 |
| d = torch.clamp(torch.abs(d1), 1e-8) * sign |
|
|
| H = src_intrs.unsqueeze(1) @ ( |
| R_rel.unsqueeze(1) + t_rel.unsqueeze(1) @ rot_normals.view(1, N_pts, 1, 3) / d.view(1, |
| N_pts, |
| 1, 1) |
| ) @ inv_ref_intr.view(1, 1, 3, 3) |
|
|
| |
| H_invalid = src_intrs.unsqueeze(1) @ ( |
| R_rel.unsqueeze(1) + t_rel.unsqueeze(1) @ self.z_axis.to(device).view(1, 1, 1, 3).expand(-1, N_pts, |
| -1, |
| -1) / sampled_dists.view( |
| 1, N_pts, 1, 1) |
| ) @ inv_ref_intr.view(1, 1, 3, 3) |
| tmp_m = ~valid_hom.view(-1, N_src).t() |
| H[tmp_m] = H_invalid[tmp_m] |
|
|
| pixels = uv.view(N_rays, 1, 2) + self.offsets.float().to(device) |
| Npx = pixels.shape[1] |
| grid, warp_mask_full = self.patch_homography(H, pixels) |
|
|
| warp_mask_full = warp_mask_full & (grid[..., 0] < (sizeW - self.h_patch_size)) & ( |
| grid[..., 1] < (sizeH - self.h_patch_size)) & (grid >= self.h_patch_size).all(dim=-1) |
| warp_mask_full = warp_mask_full.view(N_src, N_rays, n_samples, Npx) |
|
|
| grid = torch.clamp(normalize(grid, sizeH, sizeW), -10, 10) |
|
|
| sampled_rgb_val = F.grid_sample(src_imgs, grid.view(N_src, -1, 1, 2), align_corners=True).squeeze( |
| -1).transpose(1, 2) |
| sampled_rgb_val = sampled_rgb_val.view(N_src, N_rays, n_samples, Npx, 3) |
|
|
| warp_mask_full = warp_mask_full.permute(1, 2, 0, 3).contiguous() |
| sampled_rgb_val = sampled_rgb_val.permute(1, 2, 0, 3, 4).contiguous() |
|
|
| return sampled_rgb_val, warp_mask_full |
|
|
| def patch_homography(self, H, uv): |
| N, Npx = uv.shape[:2] |
| Nsrc = H.shape[0] |
| H = H.view(Nsrc, N, -1, 3, 3) |
| hom_uv = add_hom(uv) |
|
|
| |
| |
| tmp = torch.einsum("vprik,pok->vproi", H, hom_uv).reshape(Nsrc, -1, 3) |
|
|
| grid = tmp[..., :2] / torch.clamp(tmp[..., 2:], 1e-8) |
| mask = tmp[..., 2] > 0 |
| return grid, mask |
|
|
|
|
| def add_hom(pts): |
| try: |
| dev = pts.device |
| ones = torch.ones(pts.shape[:-1], device=dev).unsqueeze(-1) |
| return torch.cat((pts, ones), dim=-1) |
|
|
| except AttributeError: |
| ones = np.ones((pts.shape[0], 1)) |
| return np.concatenate((pts, ones), axis=1) |
|
|
|
|
| def normalize(flow, h, w, clamp=None): |
| |
| try: |
| h.device |
|
|
| except AttributeError: |
| h = torch.tensor(h, device=flow.device).float().unsqueeze(0) |
| w = torch.tensor(w, device=flow.device).float().unsqueeze(0) |
|
|
| if len(flow.shape) == 4: |
| w = w.unsqueeze(1).unsqueeze(2) |
| h = h.unsqueeze(1).unsqueeze(2) |
| elif len(flow.shape) == 3: |
| w = w.unsqueeze(1) |
| h = h.unsqueeze(1) |
| elif len(flow.shape) == 5: |
| w = w.unsqueeze(0).unsqueeze(2).unsqueeze(2) |
| h = h.unsqueeze(0).unsqueeze(2).unsqueeze(2) |
|
|
| res = torch.empty_like(flow) |
| if res.shape[-1] == 3: |
| res[..., 2] = 1 |
|
|
| |
| |
| res[..., 0] = 2 * flow[..., 0] / (w - 1) - 1 |
| res[..., 1] = 2 * flow[..., 1] / (h - 1) - 1 |
|
|
| if clamp: |
| return torch.clamp(res, -clamp, clamp) |
| else: |
| return res |
|
|
|
|
| def build_patch_offset(h_patch_size): |
| offsets = torch.arange(-h_patch_size, h_patch_size + 1) |
| return torch.stack(torch.meshgrid(offsets, offsets, indexing="ij")[::-1], dim=-1).view(1, -1, 2) |
|
|