Hi All,

I am looking at generating some detection proposals, see  Hosang, Jan, et al. "What makes for effective detection proposals?." arXiv preprint arXiv:1502.05082 (2015), http://arxiv.org/pdf/1502.05082.pdf    Starting with the Selective Search algorithm, Section 3 of  Uijlings, Jasper RR, et al. "Selective search for object recognition." International journal of computer vision 104.2 (2013): 154-171,    https://staff.fnwi.uva.nl/th.gevers/pub/GeversIJCV2013.pdf

The basic idea is the performing a hierarchical merging of the image, where each new merge get added to the list of regions suspected to contain an object, you can capture objects at all scales.  This reduces the search space significantly than say compared to floating window.  The output is NOT a image segmentaiton, rather a list of regions (bounding boxes) of potential objects (deteciton proposals).

I have looked in the gallery at RAG Merging http://scikit-image.org/docs/dev/auto_examples/plot_rag_merge.html,  fairly confident I can setup the callback methods to provided the similarity measure.   I am naively hoping that future.graph.hierarchical(), even though it seems to output a segmentation (labels), can be easily adapted to the task.    What would be the best way to have future.graph.merge_hierarchica() merge regions with the "highest" similarity measure, rather thana threshold?   What would be the best way future.graph.merge_hierarchica() save each merged region?   Tried setting  "in_place" to false, but didn't notice any difference.

Any help appreciated,

Brickle.
--