End To End Transformer Based Object Detection With Resnet 101 Backbone

via “modular backbone-head architecture with pluggable feature extractors”

Meta's modular object detection platform on PyTorch.

Unique: Uses a two-level registry system (@BACKBONE_REGISTRY, @ROI_HEADS_REGISTRY) with standardized FPN output contracts, allowing arbitrary backbone-head combinations without modifying model code — unlike monolithic detection frameworks where backbones and heads are tightly coupled

vs others: More composable than MMDetection because Detectron2's FPN standardization enables true plug-and-play backbone swapping; cleaner than custom PyTorch implementations because the registry pattern eliminates boilerplate instantiation code

via “single-stage detector with anchor-free and anchor-based variants”

OpenMMLab detection toolbox with 300+ models.

Unique: Provides both anchor-based (RetinaNet, ATSS) and anchor-free (FCOS, CenterNet) single-stage detectors with unified training pipeline, allowing direct comparison of approaches; uses focal loss to address class imbalance without hard negative mining, enabling end-to-end training

vs others: Faster inference than two-stage detectors (Faster R-CNN) with comparable accuracy on large objects; more flexible than YOLO because anchor aspect ratios and scales are configurable per dataset; better documented than EfficientDet with 300+ pre-trained checkpoints across architectures

via “vision transformer-based object detection with patch tokenization”

object-detection model by undefined. 7,35,352 downloads.

Unique: Uses pure Vision Transformer architecture with patch-based tokenization (no CNN backbone) for object detection, treating detection as a sequence-to-sequence task rather than region-proposal-based approach. Implements efficient attention mechanisms that scale better to high-resolution images than traditional ViT by using adaptive patch merging.

vs others: Faster inference than standard ViT-based detectors due to optimized patch tokenization, but trades accuracy for speed compared to Faster R-CNN; better suited for edge deployment than Mask R-CNN while maintaining transformer composability with language models

via “end-to-end transformer-based object detection with resnet-50 backbone”

object-detection model by undefined. 2,39,063 downloads.

Unique: DETR (Detection Transformer) eliminates hand-designed detection components (anchors, NMS) by formulating detection as a set prediction problem with bipartite matching, using a pure transformer encoder-decoder on top of ResNet-50 features rather than region proposal networks or anchor grids

vs others: Simpler architecture than Faster R-CNN (no RPN, no NMS) and more interpretable than YOLO, but slower inference and weaker small-object detection make it better suited for research and moderate-latency applications than production real-time systems

via “resnet-50 backbone feature extraction with transformer refinement”

object-detection model by undefined. 2,04,862 downloads.

Unique: Combines ImageNet-pretrained ResNet-50 CNN backbone with DETR transformer encoder-decoder, enabling both transfer learning from general vision tasks and document-specific spatial reasoning via attention, rather than using either CNN-only (Faster R-CNN) or transformer-only (ViT) approaches

vs others: More accurate than ResNet-50 alone for document tables because transformer attention captures long-range dependencies between table elements, and more efficient than pure vision transformers because ResNet-50 backbone provides strong inductive bias for local feature extraction, reducing transformer compute requirements

via “real-time object detection with transformer-based architecture”

object-detection model by undefined. 5,21,638 downloads.

Unique: Uses transformer-based detection with anchor-free, NMS-free design (RT-DETR architecture) instead of traditional Faster R-CNN/YOLO CNN pipelines; eliminates hand-crafted anchor definitions and post-processing NMS, enabling end-to-end optimization and faster convergence during training

vs others: Faster inference than DETR variants and comparable to YOLOv8 while maintaining transformer interpretability; outperforms ResNet-50 Faster R-CNN on COCO at similar latency due to efficient attention mechanisms

via “end-to-end transformer-based object detection with resnet-101 backbone”

object-detection model by undefined. 63,737 downloads.

Unique: Uses transformer encoder-decoder with bipartite matching loss instead of anchor-based region proposals or sliding windows, eliminating hand-crafted NMS and enabling direct set prediction of objects as a sequence-to-sequence problem

vs others: Simpler pipeline than Faster R-CNN (no RPN, no NMS) and more interpretable than YOLO, but slower inference due to transformer quadratic complexity compared to single-stage detectors

via “vision transformer-based object detection with attention-weighted region proposals”

object-detection model by undefined. 83,525 downloads.

Unique: Applies pure transformer architecture (DETR-style with learnable object queries) to object detection instead of CNN backbones, enabling attention-based spatial reasoning without region proposal networks; tiny variant achieves 5.4M parameters through aggressive model compression while maintaining COCO detection capability

vs others: Simpler architecture than Faster R-CNN (no RPN) and more parameter-efficient than standard ViT detectors, but slower inference than optimized YOLO v5/v8 on edge devices due to transformer computational overhead

via “real-time object detection with transformer-based architecture”

object-detection model by undefined. 1,21,720 downloads.

Unique: Uses transformer encoder-decoder architecture with direct set prediction (eliminating anchor boxes and NMS) combined with ResNet-101-VD backbone, achieving real-time performance through efficient attention mechanisms and hybrid CNN-transformer design that balances speed and accuracy across 365 object categories from Objects365 dataset

vs others: Faster than traditional Faster R-CNN/Mask R-CNN detectors (50-100ms vs 200-400ms) while maintaining higher accuracy than lightweight YOLO variants through transformer attention, and more practical for production than ViT-based detectors due to optimized backbone selection

via “real-time object detection with transformer-based architecture”

object-detection model by undefined. 80,830 downloads.

Unique: Uses transformer encoder-decoder architecture with deformable attention mechanisms instead of traditional CNN-based region proposal networks; eliminates anchor boxes and NMS post-processing, reducing inference pipeline complexity while maintaining real-time performance through efficient attention computation

vs others: Faster inference than Faster R-CNN (no RPN overhead) and simpler than YOLO (no anchor engineering), while maintaining transformer-based reasoning for improved generalization across diverse object scales and aspect ratios

via “real-time object detection with deformable transformer attention”

object-detection model by undefined. 1,06,918 downloads.

Unique: Uses deformable transformer attention (sampling only task-relevant spatial regions) combined with ResNet-18 backbone for real-time inference, whereas standard DETR processes full feature maps with quadratic attention complexity. This architectural choice reduces FLOPs by ~40% compared to vanilla transformer detectors while maintaining anchor-free detection paradigm.

vs others: Faster than YOLOv8 on edge devices due to deformable attention efficiency, and more accurate than lightweight anchor-based detectors (MobileNet-SSD) because transformer attention captures long-range spatial relationships without hand-crafted anchor priors.

via “real-time object detection with deformable transformer architecture”

object-detection model by undefined. 32,868 downloads.

Unique: Uses deformable cross-attention instead of standard multi-head attention, allowing the model to dynamically sample only task-relevant spatial regions; combined with ResNet-50-VD backbone (a more efficient variant than standard ResNet-50), this achieves <100ms inference while maintaining COCO AP of 53.0+ without NMS post-processing

vs others: Faster inference than YOLOv8 on equivalent hardware (deformable attention vs dense convolution) and more accurate than EfficientDet-D0 on COCO while using fewer parameters than Faster R-CNN variants

via “object detection with transformer architecture”

object-detection model by undefined. 38,839 downloads.

Unique: Utilizes a unique end-to-end transformer architecture that eliminates the need for anchor boxes, making it simpler and more efficient for training.

vs others: More straightforward to implement and train compared to traditional object detection models like Faster R-CNN, which require complex anchor box configurations.

via “transformer-based detector implementation (detr, deformable detr, dino variants)”

OpenMMLab Detection Toolbox and Benchmark

Unique: Implements transformer-based detection as a set prediction problem with learnable query embeddings refined through multi-layer transformer decoders, and supports deformable attention that learns spatial offsets to focus on relevant regions, enabling efficient processing of multi-scale features without hand-crafted anchors

vs others: More efficient than vanilla DETR because deformable attention reduces computational complexity from O(n²) to O(n) by attending only to relevant spatial regions; more integrated than standalone DETR implementations because it shares backbone/neck infrastructure with CNN-based detectors, enabling easy comparison

via “object detection and instance segmentation with convolutional architectures”


Unique: Provides fastai wrappers around Faster R-CNN and Mask R-CNN that simplify the two-stage detection pipeline, handling region proposal generation, anchor matching, and loss computation automatically. Includes utilities for converting between annotation formats and visualizing predictions with bounding boxes and masks.

vs others: Faster to prototype object detection systems than implementing Faster R-CNN from scratch in PyTorch; includes pre-trained backbones (ResNet, EfficientNet) for transfer learning on custom datasets.

via “coco-object-detection-backbone-integration”

* ⭐ 01/2022: [Patches Are All You Need (ConvMixer)](https://arxiv.org/abs/2201.09792)

Unique: Achieves COCO detection performance that outperforms Swin Transformer while maintaining pure convolutional architecture, demonstrating that modernized ConvNets can compete with transformer-based backbones on detection tasks without attention mechanisms

vs others: Outperforms Swin Transformer on COCO object detection while providing simpler architecture, lower inference latency (unquantified), and better interpretability than attention-based backbones