Python cv2.COLORMAP_BONE Examples

The following are 4 code examples of cv2.COLORMAP_BONE(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module cv2 , or try the search function .
Example #1
Source File: utils.py    From kaggle_carvana_segmentation with MIT License 5 votes vote down vote up 
def heatmap(map):
    map = (map*255).astype(np.uint8)
    return cv2.applyColorMap(map, cv2.COLORMAP_BONE)
Example #2
Source File: util.py    From deconvolution with GNU General Public License v3.0 5 votes vote down vote up 
def tensor2array(tensor, max_value=None, colormap='rainbow'):
    if max_value is None:
        tensor=(tensor-tensor.min())/(tensor.max()-tensor.min()+1e-6)
        max_value = tensor.max().item()
    if tensor.ndimension() == 2 or tensor.size(0) == 1:
        try:
            import cv2
            if cv2.__version__.startswith('3'):
                color_cvt = cv2.COLOR_BGR2RGB
            else:  # 2.4
                color_cvt = cv2.cv.CV_BGR2RGB
            if colormap == 'rainbow':
                colormap = cv2.COLORMAP_RAINBOW
            elif colormap == 'bone':
                colormap = cv2.COLORMAP_BONE
            array = (tensor.squeeze().numpy()*255./max_value).clip(0, 255).astype(np.uint8)
            colored_array = cv2.applyColorMap(array, colormap)
            array = cv2.cvtColor(colored_array, color_cvt).astype(np.float32)/255
        except ImportError:
            if tensor.ndimension() == 2:
                tensor.unsqueeze_(2)
            array = (tensor.expand(tensor.size(0), tensor.size(1), 3).numpy()/max_value).clip(0,1)

    elif tensor.ndimension() == 3:
        assert(tensor.size(0) == 3)
        array = 0.5 + tensor.numpy().transpose(1, 2, 0)*0.5

    #for tensorboardx 1.4
    #array=array.transpose(2,0,1)

    return array
Example #3
Source File: utils.py    From DPSNet with MIT License 5 votes vote down vote up 
def tensor2array(tensor, max_value=255, colormap='rainbow'):
    if max_value is None:
        max_value = tensor.max()
    if tensor.ndimension() == 2 or tensor.size(0) == 1:
        try:
            import cv2
            if cv2.__version__.startswith('2'): # 2.4
                color_cvt = cv2.cv.CV_BGR2RGB
            else:  
                color_cvt = cv2.COLOR_BGR2RGB
            if colormap == 'rainbow':
                colormap = cv2.COLORMAP_RAINBOW
            elif colormap == 'bone':
                colormap = cv2.COLORMAP_BONE
            array = (255*tensor.squeeze().numpy()/max_value).clip(0, 255).astype(np.uint8)
            colored_array = cv2.applyColorMap(array, colormap)
            array = cv2.cvtColor(colored_array, color_cvt).astype(np.float32)/255
            #array = array.transpose(2, 0, 1)
        except ImportError:
            if tensor.ndimension() == 2:
                tensor.unsqueeze_(2)
            array = (tensor.expand(tensor.size(0), tensor.size(1), 3).numpy()/max_value).clip(0,1)

    elif tensor.ndimension() == 3:
        #assert(tensor.size(0) == 3)
        #array = 0.5 + tensor.numpy()*0.5
        array = 0.5 + tensor.numpy().transpose(1,2,0)*0.5
    return array
Example #4
Source File: visualizer.py    From hd3 with BSD 3-Clause "New" or "Revised" License 4 votes vote down vote up 
def get_visualization(img_list, label_list, ms_vect, ms_prob, ds=6, idx=0):
    dim = ms_vect[0].size(1)
    H, W = img_list[0].size()[2:]
    with torch.no_grad():
        raw_img0 = _recover_img(img_list[0][idx].data)
        raw_img1 = _recover_img(img_list[1][idx].data)
        for l in range(len(ms_vect)):
            # image
            vis_list = [raw_img0, raw_img1]

            # ground-truth flow
            gt_flo, valid_mask = downsample_flow(label_list[0],
                                                 1 / 2**(ds - l))
            gt_flo = F.interpolate(gt_flo, (H, W), mode='nearest')[idx]
            valid_mask = F.interpolate(valid_mask, (H, W), mode='nearest')[idx]
            max_mag1 = torch.max(torch.norm(gt_flo, 2, 0))

            # predicted flow
            pred_flo = ms_vect[l]
            if dim == 1:
                pred_flo = disp2flow(pred_flo)
            pred_flo = F.interpolate(pred_flo, (H, W), mode='nearest')[idx]
            max_mag2 = torch.max(torch.norm(pred_flo, 2, 0))

            max_mag = max(float(max_mag1), float(max_mag2))
            vis_list.append(_flow_to_img(gt_flo, max_mag))
            vis_list.append(_flow_to_img(pred_flo, max_mag))

            # epe error visualization
            epe_error = torch.norm(
                pred_flo - gt_flo, 2, 0, keepdim=False) * valid_mask[0, :, :]
            normalizer = max(torch.max(epe_error), 1)
            epe_error = 1 - epe_error / normalizer
            vis_list.append(_visualize_heat(epe_error))

            # confidence map visualization
            prob = ms_prob[l].data
            prob = prob_gather(prob, normalize=True, dim=dim)
            if prob.size(2) != H or prob.size(3) != W:
                prob = F.interpolate(prob, (H, W), mode='nearest')
            vis_list.append(
                _visualize_heat(prob[idx].squeeze(), cv2.COLORMAP_BONE))

            vis = torch.cat(vis_list, dim=2)
            if l == 0:
                ms_vis = vis
            else:
                ms_vis = torch.cat([ms_vis, vis], dim=1)

        return ms_vis.unsqueeze(0)