Focal Loss for Dense Object Detection
Tsung-Yi Lin, Priya Goyal, +3
Separate alpha weighting from focal modulation, derive the complete logit gradient, and inspect population imbalance, prior initialization and RetinaNet evidence.
89 reviews with explanations and code.Machine learning research explained through diagrams, mathematics and code.
Browse by research area
Showing 10-18 of 89 papers
Tsung-Yi Lin, Priya Goyal, +3
Separate alpha weighting from focal modulation, derive the complete logit gradient, and inspect population imbalance, prior initialization and RetinaNet evidence.
FPN
Tsung-Yi Lin, Piotr Dollár, +4
Build FPN from C2–C5, lateral 1×1 projections, nearest top-down addition and 3×3 outputs; trace RoI level assignment and the original detection evidence.
Yann LeCun, Léon Bottou, +2
Read the original LeNet-5 through shared filters, learned subsampling, partial C3 connectivity and fixed RBF outputs, with executable references and the 1998 document-recognition evidence.
Sepp Hochreiter, Jürgen Schmidhuber
Explore the original no-forget-gate LSTM: gated writes and reads, the constant error carousel, path derivatives and the delayed-sequence experiments behind the 1997 paper.
Kaiming He, Georgia Gkioxari, +2
Trace continuous RoIAlign sample coordinates, bilinear interpolation and matched-class mask loss, then examine the original Mask R-CNN evidence and limits.
David G. Lowe
Trace SIFT from extrema in scale space to orientation-normalized 128-dimensional descriptors and ambiguity-aware matching, with original PDF excerpts and inspectable reference operations.
Olaf Ronneberger, Philipp Fischer, +1
Reconstruct the original 572-to-388 valid-convolution U-Net, crop-and-concatenate skips, overlap tiling and boundary-weighted training, with executable geometry.
Karen Simonyan, Andrew Zisserman
Explore VGG’s 3×3 stacks, exact VGG-16/VGG-19 shapes and parameter counts, then separate architecture effects from scale, crop and ensemble evaluation in the original study.
Yansong Peng, Hebei Li, +4
Explore D-FINE’s edge distributions, iterative refinement, localization loss and self-distillation with interactive figures, Python and benchmark caveats.