Trackpy tp.batch works but annotate.3d gives attribute error related to image dimension?

I have a tiff stack from a video that I created in Fiji. I want to track particles and follow their intensity with time.

I followed the walkthrough for Trackpy and it seems to work well for tp.batch, however when I try to visualize the results with annotate 3d, I get an error:

if image.ndim != 3 and not (image.ndim == 4 and image.shape[-1] in (3, 4)):

  File "PATH", line 256, in __getattr__
    raise AttributeError

AttributeError

Any idea what could cause this?


Here are the results from tp.batch with head()

Frame 5: 1 features
Frame 6: 1 features
Frame 7: 1 features
Frame 8: 1 features
Frame 9: 1 features
Frame 10: 1 features
Frame 11: 2 features
Frame 12: 2 features
Frame 13: 2 features
Frame 14: 2 features
Out[77]: 
          y           x          mass  ...      ep_y      ep_x  frame
0  9.478312  642.114524  11851.919420  ...  0.021276  0.085622      5
1  9.298654  608.647362  10689.680487  ...  0.024296  0.097777      6
2  9.034290  575.615509  10385.377583  ...  0.025193  0.101387      7
3  9.060161  541.555573  10356.535314  ...  0.025626  0.103128      8
4  8.994890  507.360459   9833.859577  ...  0.029229  0.117628      9

Here is the information about the tiff from FIJI

Title: stack4.tif
Width:  1330 pixels
Height:  24 pixels
Depth:  301 pixels
Size:  9.2MB
Voxel size: 1x1x1 pixel^3
ID: -4
Bits per pixel: 8 (grayscale LUT)
Display range: 0-255
Frame: 1/301 (slice:600)
Frame rate: 5.88235 fps
Frame interval: 0.01700 sec
No threshold
ScaleToFit: false
Uncalibrated
Path: PATH/stack4.tif
Screen location: 100,245 (1440x900)
SetMenuBarCount: 5 (0ms)
Coordinate origin:  0,0,0
No properties
No overlay
No selection

Here is my code

import pims
import trackpy as tp

frames=pims.open('PATH/stack4.tif');
plt.imshow(frames[10]);
plt.show()
f = tp.locate(frames[10], (11,39), threshold=75, minmass=100, separation=5);
tp.annotate(f, frames[10]);
f = tp.batch(frames[5:15], (11,39), threshold=75, minmass=100, separation=5);
f.head()
tp.annotate3d(f, frames[5:15]);