Speed/accuracy trade-offs for modern convolutional object detectors · HackerPapers discussions

Speed/accuracy trade-offs for modern convolutional object detectors — mentioned by mendeza

Wed, 06 Nov 2019 16:50:40 GMT

Also, what is a hard but an important question to figure out is how well are these models are preforming, and will they perform well to your problem. This paper gives a lot of insight into how image size and other factors really impact performance of modern object detection: https://arxiv.org/pdf/1611.10012.pdf. How large or small your object is in your dataset is also something to consider since there is a limit to how small of an object your detector can recognize. Also, I would recommend this repo to better understand how object detection models are evaluated as well…

Paper and discussion history. A mention is not an endorsement.

Speed/accuracy trade-offs for modern convolutional object detectors — mentioned by hedgehog

Mon, 21 Aug 2017 06:23:51 GMT

I missed that the site was updated for YOLO9000, it is very impressive. The Google paper is recent though (July), maybe you're thinking of https://arxiv.org/abs/1611.10012? Anyway, what I meant to point out was that there has been a lot of work since on improving the efficiency of the architectures used for feature extraction (Xception, MobileNet, ShuffleNet, NASNet).

Paper and discussion history. A mention is not an endorsement.

Speed/accuracy trade-offs for modern convolutional object detectors — mentioned by pveierland

Sat, 17 Jun 2017 10:51:36 GMT

The researchers have created a framework for object detection such that one can easily experiment with using different feature extraction networks, separated from the "meta-architecture" such as Faster R-CNN, R-FCN, or SSD, used to handle the object detection task. They compare many models using this framework, described in https://arxiv.org/abs/1611.10012 - and they were able to construct the winning entry of the COCO 2016 detection challenge based on this research.

Paper and discussion history. A mention is not an endorsement.