Segmentation results on ShapeNet part benchmark

[TheJasal]

****** Airplane: 0.001279 ****** Bag: 0.125000 ****** Cap: 0.000000 ****** Car: 0.015432 ****** Chair: 0.066288 ****** Earphone: 0.222222 ****** Guitar: 0.000000 ****** Knife: 0.000000 ****** Lamp: 0.204545 ****** Laptop: 0.000000 ****** Motorbike: 0.102564 ****** Mug: 0.000000 ****** Pistol: 0.000000 ****** Rocket: 0.041667 ****** Skateboard: 0.066667 ****** Table: 0.159297 ************ Test Loss: 3.403158 ************ Class_mIoU: 0.062810 ************ Instance_mIoU: 0.084091 @Yochengliu
why the model of Cap, Mug and Pistol instance_ious is always 0.000000 ?

[tn00364361]

I had the same issue with Pytorch 0.4. I manage to solve it by modifying line 103 to 114 in voting_evaluate_partseg.py from

for b in range(len(cls)):
    segp = pred_val[b, :]
    segl = target[b, :]
    cat = seg_label_to_cat[segl[0]]
    part_ious = [0.0 for _ in range(len(seg_classes[cat]))]
    for l in seg_classes[cat]:
        if torch.sum((segl == l) | (segp == l)) == 0:
            # part is not present in this shape
            part_ious[l - seg_classes[cat][0]] = 1.0
        else:
            part_ious[l - seg_classes[cat][0]] = torch.sum((segl == l) & (segp == l)) / float(torch.sum((segl == l) | (segp == l)))
    shape_ious[cat].append(np.mean(part_ious))

to

for b in range(len(cls)):
    segp = pred_val[b, :].numpy()
    segl = target[b, :].numpy()
    cat = seg_label_to_cat[segl[0]]
    part_ious = [0.0 for _ in range(len(seg_classes[cat]))]
    for l in seg_classes[cat]:
        if np.sum((segl == l) | (segp == l)) == 0:
            # part is not present in this shape
            part_ious[l - seg_classes[cat][0]] = 1.0
        else:
            part_ious[l - seg_classes[cat][0]] = np.sum((segl == l) & (segp == l)) / float(np.sum((segl == l) | (segp == l)))
    shape_ious[cat].append(np.mean(part_ious))

I also did the same mod in train_partseg.py.

[Jamesli0123]

I had the same issue with Pytorch 0.4. I manage to solve it by modifying line 103 to 114 in voting_evaluate_partseg.py from

for b in range(len(cls)):
    segp = pred_val[b, :]
    segl = target[b, :]
    cat = seg_label_to_cat[segl[0]]
    part_ious = [0.0 for _ in range(len(seg_classes[cat]))]
    for l in seg_classes[cat]:
        if torch.sum((segl == l) | (segp == l)) == 0:
            # part is not present in this shape
            part_ious[l - seg_classes[cat][0]] = 1.0
        else:
            part_ious[l - seg_classes[cat][0]] = torch.sum((segl == l) & (segp == l)) / float(torch.sum((segl == l) | (segp == l)))
    shape_ious[cat].append(np.mean(part_ious))

to

for b in range(len(cls)):
    segp = pred_val[b, :].numpy()
    segl = target[b, :].numpy()
    cat = seg_label_to_cat[segl[0]]
    part_ious = [0.0 for _ in range(len(seg_classes[cat]))]
    for l in seg_classes[cat]:
        if np.sum((segl == l) | (segp == l)) == 0:
            # part is not present in this shape
            part_ious[l - seg_classes[cat][0]] = 1.0
        else:
            part_ious[l - seg_classes[cat][0]] = np.sum((segl == l) & (segp == l)) / float(np.sum((segl == l) | (segp == l)))
    shape_ious[cat].append(np.mean(part_ious))

I also did the same mod in train_partseg.py. i modified it according to your method and then used the model provided by the author. The training result is normal