diff --git a/esphome/components/api/api.proto b/esphome/components/api/api.proto
index 684540ffa6..e625c12ffd 100644
--- a/esphome/components/api/api.proto
+++ b/esphome/components/api/api.proto
@@ -763,7 +763,7 @@ message ExecuteServiceRequest {
 message ListEntitiesCameraResponse {
   option (id) = 43;
   option (source) = SOURCE_SERVER;
-  option (ifdef) = "USE_ESP32_CAMERA";
+  option (ifdef) = "USE_CAMERA";
 
   string object_id = 1;
   fixed32 key = 2;
@@ -777,7 +777,7 @@ message ListEntitiesCameraResponse {
 message CameraImageResponse {
   option (id) = 44;
   option (source) = SOURCE_SERVER;
-  option (ifdef) = "USE_ESP32_CAMERA";
+  option (ifdef) = "USE_CAMERA";
 
   fixed32 key = 1;
   bytes data = 2;
@@ -786,7 +786,7 @@ message CameraImageResponse {
 message CameraImageRequest {
   option (id) = 45;
   option (source) = SOURCE_CLIENT;
-  option (ifdef) = "USE_ESP32_CAMERA";
+  option (ifdef) = "USE_CAMERA";
   option (no_delay) = true;
 
   bool single = 1;
diff --git a/esphome/components/api/api_connection.cpp b/esphome/components/api/api_connection.cpp
index bb55a2ccf6..6d26af714e 100644
--- a/esphome/components/api/api_connection.cpp
+++ b/esphome/components/api/api_connection.cpp
@@ -39,6 +39,11 @@ APIConnection::APIConnection(std::unique_ptr<socket::Socket> sock, APIServer *pa
 #else
 #error "No frame helper defined"
 #endif
+
+#ifdef USE_CAMERA
+  if (camera::Camera::global_camera != nullptr)
+    this->image_reader_ = std::unique_ptr<camera::CameraImageReader>{camera::Camera::global_camera->create_image_reader()};
+#endif
 }
 void APIConnection::start() {
   this->last_traffic_ = millis();
@@ -149,24 +154,24 @@ void APIConnection::loop() {
     }
   }
 
-#ifdef USE_ESP32_CAMERA
-  if (this->image_reader_.available() && this->helper_->can_write_without_blocking()) {
-    uint32_t to_send = std::min((size_t) 1024, this->image_reader_.available());
+#ifdef USE_CAMERA
+  if (this->image_reader_ && this->image_reader_->available() && this->helper_->can_write_without_blocking()) {
+    uint32_t to_send = std::min((size_t) 1024, this->image_reader_->available());
     auto buffer = this->create_buffer();
     // fixed32 key = 1;
-    buffer.encode_fixed32(1, esp32_camera::global_esp32_camera->get_object_id_hash());
+    buffer.encode_fixed32(1, camera::Camera::global_camera->get_object_id_hash());
     // bytes data = 2;
-    buffer.encode_bytes(2, this->image_reader_.peek_data_buffer(), to_send);
+    buffer.encode_bytes(2, this->image_reader_->peek_data_buffer(), to_send);
     // bool done = 3;
-    bool done = this->image_reader_.available() == to_send;
+    bool done = this->image_reader_->available() == to_send;
     buffer.encode_bool(3, done);
     bool success = this->send_buffer(buffer, 44);
 
     if (success) {
-      this->image_reader_.consume_data(to_send);
+      this->image_reader_->consume_data(to_send);
     }
     if (success && done) {
-      this->image_reader_.return_image();
+      this->image_reader_->return_image();
     }
   }
 #endif
@@ -1061,17 +1066,19 @@ void APIConnection::media_player_command(const MediaPlayerCommandRequest &msg) {
 }
 #endif
 
-#ifdef USE_ESP32_CAMERA
-void APIConnection::send_camera_state(std::shared_ptr<esp32_camera::CameraImage> image) {
+#ifdef USE_CAMERA
+void APIConnection::send_camera_state(std::shared_ptr<camera::CameraImage> image) {
   if (!this->state_subscription_)
     return;
-  if (this->image_reader_.available())
+  if (!this->image_reader_)
     return;
-  if (image->was_requested_by(esphome::esp32_camera::API_REQUESTER) ||
-      image->was_requested_by(esphome::esp32_camera::IDLE))
-    this->image_reader_.set_image(std::move(image));
+  if (this->image_reader_->available())
+    return;
+  if (image->was_requested_by(esphome::camera::API_REQUESTER) ||
+      image->was_requested_by(esphome::camera::IDLE))
+    this->image_reader_->set_image(std::move(image));
 }
-bool APIConnection::send_camera_info(esp32_camera::ESP32Camera *camera) {
+bool APIConnection::send_camera_info(camera::Camera *camera) {
   ListEntitiesCameraResponse msg;
   msg.key = camera->get_object_id_hash();
   msg.object_id = camera->get_object_id();
@@ -1084,16 +1091,16 @@ bool APIConnection::send_camera_info(esp32_camera::ESP32Camera *camera) {
   return this->send_list_entities_camera_response(msg);
 }
 void APIConnection::camera_image(const CameraImageRequest &msg) {
-  if (esp32_camera::global_esp32_camera == nullptr)
+  if (camera::Camera::global_camera == nullptr)
     return;
 
   if (msg.single)
-    esp32_camera::global_esp32_camera->request_image(esphome::esp32_camera::API_REQUESTER);
+    camera::Camera::global_camera->request_image(esphome::camera::API_REQUESTER);
   if (msg.stream) {
-    esp32_camera::global_esp32_camera->start_stream(esphome::esp32_camera::API_REQUESTER);
+    camera::Camera::global_camera->start_stream(esphome::camera::API_REQUESTER);
 
     App.scheduler.set_timeout(this->parent_, "api_esp32_camera_stop_stream", ESP32_CAMERA_STOP_STREAM, []() {
-      esp32_camera::global_esp32_camera->stop_stream(esphome::esp32_camera::API_REQUESTER);
+      camera::Camera::global_camera->stop_stream(esphome::camera::API_REQUESTER);
     });
   }
 }
diff --git a/esphome/components/api/api_connection.h b/esphome/components/api/api_connection.h
index 043aaee421..77a5b20b29 100644
--- a/esphome/components/api/api_connection.h
+++ b/esphome/components/api/api_connection.h
@@ -58,9 +58,9 @@ class APIConnection : public APIServerConnection {
   bool send_text_sensor_state(text_sensor::TextSensor *text_sensor, std::string state);
   bool send_text_sensor_info(text_sensor::TextSensor *text_sensor);
 #endif
-#ifdef USE_ESP32_CAMERA
-  void send_camera_state(std::shared_ptr<esp32_camera::CameraImage> image);
-  bool send_camera_info(esp32_camera::ESP32Camera *camera);
+#ifdef USE_CAMERA
+  void send_camera_state(std::shared_ptr<camera::CameraImage> image);
+  bool send_camera_info(camera::Camera *camera);
   void camera_image(const CameraImageRequest &msg) override;
 #endif
 #ifdef USE_CLIMATE
@@ -249,8 +249,8 @@ class APIConnection : public APIServerConnection {
   std::string client_combined_info_;
   uint32_t client_api_version_major_{0};
   uint32_t client_api_version_minor_{0};
-#ifdef USE_ESP32_CAMERA
-  esp32_camera::CameraImageReader image_reader_;
+#ifdef USE_CAMERA
+  std::unique_ptr<camera::CameraImageReader> image_reader_;
 #endif
 
   bool state_subscription_{false};
diff --git a/esphome/components/api/api_pb2_service.cpp b/esphome/components/api/api_pb2_service.cpp
index 6e11d7169d..234b96e4d0 100644
--- a/esphome/components/api/api_pb2_service.cpp
+++ b/esphome/components/api/api_pb2_service.cpp
@@ -210,7 +210,8 @@ bool APIServerConnectionBase::send_list_entities_services_response(const ListEnt
 #endif
   return this->send_message_<ListEntitiesServicesResponse>(msg, 41);
 }
-#ifdef USE_ESP32_CAMERA
+
+#ifdef USE_CAMERA
 bool APIServerConnectionBase::send_list_entities_camera_response(const ListEntitiesCameraResponse &msg) {
 #ifdef HAS_PROTO_MESSAGE_DUMP
   ESP_LOGVV(TAG, "send_list_entities_camera_response: %s", msg.dump().c_str());
@@ -218,7 +219,7 @@ bool APIServerConnectionBase::send_list_entities_camera_response(const ListEntit
   return this->send_message_<ListEntitiesCameraResponse>(msg, 43);
 }
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
 bool APIServerConnectionBase::send_camera_image_response(const CameraImageResponse &msg) {
 #ifdef HAS_PROTO_MESSAGE_DUMP
   ESP_LOGVV(TAG, "send_camera_image_response: %s", msg.dump().c_str());
@@ -226,8 +227,6 @@ bool APIServerConnectionBase::send_camera_image_response(const CameraImageRespon
   return this->send_message_<CameraImageResponse>(msg, 44);
 }
 #endif
-#ifdef USE_ESP32_CAMERA
-#endif
 #ifdef USE_CLIMATE
 bool APIServerConnectionBase::send_list_entities_climate_response(const ListEntitiesClimateResponse &msg) {
 #ifdef HAS_PROTO_MESSAGE_DUMP
@@ -843,7 +842,7 @@ bool APIServerConnectionBase::read_message(uint32_t msg_size, uint32_t msg_type,
       break;
     }
     case 45: {
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
       CameraImageRequest msg;
       msg.decode(msg_data, msg_size);
 #ifdef HAS_PROTO_MESSAGE_DUMP
@@ -1363,7 +1362,7 @@ void APIServerConnection::on_switch_command_request(const SwitchCommandRequest &
   this->switch_command(msg);
 }
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
 void APIServerConnection::on_camera_image_request(const CameraImageRequest &msg) {
   if (!this->is_connection_setup()) {
     this->on_no_setup_connection();
diff --git a/esphome/components/api/api_pb2_service.h b/esphome/components/api/api_pb2_service.h
index 51b94bf530..1c424b0582 100644
--- a/esphome/components/api/api_pb2_service.h
+++ b/esphome/components/api/api_pb2_service.h
@@ -94,13 +94,13 @@ class APIServerConnectionBase : public ProtoService {
   virtual void on_get_time_response(const GetTimeResponse &value){};
   bool send_list_entities_services_response(const ListEntitiesServicesResponse &msg);
   virtual void on_execute_service_request(const ExecuteServiceRequest &value){};
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
   bool send_list_entities_camera_response(const ListEntitiesCameraResponse &msg);
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
   bool send_camera_image_response(const CameraImageResponse &msg);
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
   virtual void on_camera_image_request(const CameraImageRequest &value){};
 #endif
 #ifdef USE_CLIMATE
@@ -361,7 +361,7 @@ class APIServerConnection : public APIServerConnectionBase {
 #ifdef USE_SWITCH
   virtual void switch_command(const SwitchCommandRequest &msg) = 0;
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
   virtual void camera_image(const CameraImageRequest &msg) = 0;
 #endif
 #ifdef USE_CLIMATE
@@ -469,7 +469,7 @@ class APIServerConnection : public APIServerConnectionBase {
 #ifdef USE_SWITCH
   void on_switch_command_request(const SwitchCommandRequest &msg) override;
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
   void on_camera_image_request(const CameraImageRequest &msg) override;
 #endif
 #ifdef USE_CLIMATE
diff --git a/esphome/components/api/api_server.cpp b/esphome/components/api/api_server.cpp
index f16b5a13cf..b353143e42 100644
--- a/esphome/components/api/api_server.cpp
+++ b/esphome/components/api/api_server.cpp
@@ -80,10 +80,10 @@ void APIServer::setup() {
 
   this->last_connected_ = millis();
 
-#ifdef USE_ESP32_CAMERA
-  if (esp32_camera::global_esp32_camera != nullptr && !esp32_camera::global_esp32_camera->is_internal()) {
-    esp32_camera::global_esp32_camera->add_image_callback(
-        [this](const std::shared_ptr<esp32_camera::CameraImage> &image) {
+#ifdef USE_CAMERA
+  if (camera::Camera::global_camera != nullptr && !camera::Camera::global_camera->is_internal()) {
+    camera::Camera::global_camera->add_image_callback(
+        [this](const std::shared_ptr<camera::CameraImage> &image) {
           for (auto &c : this->clients_) {
             if (!c->remove_)
               c->send_camera_state(image);
diff --git a/esphome/components/api/list_entities.cpp b/esphome/components/api/list_entities.cpp
index 8540c5cee8..0be0b84745 100644
--- a/esphome/components/api/list_entities.cpp
+++ b/esphome/components/api/list_entities.cpp
@@ -50,8 +50,8 @@ bool ListEntitiesIterator::on_service(UserServiceDescriptor *service) {
   return this->client_->send_list_entities_services_response(resp);
 }
 
-#ifdef USE_ESP32_CAMERA
-bool ListEntitiesIterator::on_camera(esp32_camera::ESP32Camera *camera) {
+#ifdef USE_CAMERA
+bool ListEntitiesIterator::on_camera(camera::Camera *camera) {
   return this->client_->send_camera_info(camera);
 }
 #endif
diff --git a/esphome/components/api/list_entities.h b/esphome/components/api/list_entities.h
index 1821bbee4c..689789410f 100644
--- a/esphome/components/api/list_entities.h
+++ b/esphome/components/api/list_entities.h
@@ -37,8 +37,8 @@ class ListEntitiesIterator : public ComponentIterator {
   bool on_text_sensor(text_sensor::TextSensor *text_sensor) override;
 #endif
   bool on_service(UserServiceDescriptor *service) override;
-#ifdef USE_ESP32_CAMERA
-  bool on_camera(esp32_camera::ESP32Camera *camera) override;
+#ifdef USE_CAMERA
+  bool on_camera(camera::Camera *camera) override;
 #endif
 #ifdef USE_CLIMATE
   bool on_climate(climate::Climate *climate) override;
diff --git a/esphome/components/camera/__init__.py b/esphome/components/camera/__init__.py
new file mode 100644
index 0000000000..7c887793e3
--- /dev/null
+++ b/esphome/components/camera/__init__.py
@@ -0,0 +1,7 @@
+import esphome.codegen as cg
+import esphome.config_validation as cv
+
+CONFIG_SCHEMA = cv.Schema({})
+
+async def to_code(config):
+    cg.add_define("USE_CAMERA")
diff --git a/esphome/components/camera/camera.cpp b/esphome/components/camera/camera.cpp
new file mode 100644
index 0000000000..680f786d3a
--- /dev/null
+++ b/esphome/components/camera/camera.cpp
@@ -0,0 +1,13 @@
+#include "camera.h"
+
+namespace esphome {
+namespace camera {
+
+Camera *Camera::global_camera = nullptr;
+
+Camera::Camera() {
+  global_camera = this;
+}
+
+}  // namespace camera
+}  // namespace esphome
diff --git a/esphome/components/camera/camera.h b/esphome/components/camera/camera.h
new file mode 100644
index 0000000000..4113d456ef
--- /dev/null
+++ b/esphome/components/camera/camera.h
@@ -0,0 +1,76 @@
+#pragma once
+
+#include "esphome/core/automation.h"
+#include "esphome/core/component.h"
+#include "esphome/core/entity_base.h"
+#include "esphome/core/helpers.h"
+
+namespace esphome {
+namespace camera {
+
+/** Different sources for filtering.
+ *  IDLE: Camera requests to send an image to the API.
+ *  API_REQUESTER: API requests a new image.
+ *  WEB_REQUESTER: ESP32 web server request an image. Ignored by API.
+ */
+enum CameraRequester { IDLE, API_REQUESTER, WEB_REQUESTER };
+
+/** Abstract camera image base class.
+ *  Encapsulates the JPEG encoded data and it is shared among
+ *  all connected clients.
+ */
+class CameraImage {
+ public:
+  virtual uint8_t *get_data_buffer() = 0;
+  virtual size_t get_data_length() = 0;
+  virtual bool was_requested_by(CameraRequester requester) const = 0;
+  virtual ~CameraImage() {}
+};
+
+/** Abstract image reader base class.
+ *  Keeps track of the data offset of the camera image and
+ *  how many bytes are remaining to read. When the image
+ *  is returned, the shared_ptr is reset and the camera can
+ *  reuse the memory of the camera image.
+ */
+class CameraImageReader {
+ public:
+  virtual void set_image(std::shared_ptr<CameraImage> image) = 0;
+  virtual size_t available() const = 0;
+  virtual uint8_t *peek_data_buffer() = 0;
+  virtual void consume_data(size_t consumed) = 0;
+  virtual void return_image() = 0;
+  virtual ~CameraImageReader() {}
+};
+
+/** Abstract camera base class. Collaborates with API.
+ *  1) API server starts and installs callback (add_image_callback)
+ *     which is called by the camera when a new image is available.
+ *  2) New API client connects and creates a new image reader (create_image_reader).
+ *  3) API connection receives protobuf CameraImageRequest and calls request_image.
+ *  3.a) API connection receives protobuf CameraImageRequest and calls start_stream.
+ *  4) Camera implementation provides JPEG data in the CameraImage and calls callback.
+ *  5) API connection sets the image in the image reader.
+ *  6) API connection consumes data from the image reader and returns the image when finished.
+ *  7.a) Camera caputes new image and continues with 4) until start_stream is called.
+ */
+class Camera : public EntityBase {
+ public:
+  Camera();
+  // Camera implementation invokes callback to publish a new image.
+  virtual void add_image_callback(std::function<void(std::shared_ptr<CameraImage>)> &&callback) = 0;
+  /// Returns a new camera image reader that keeps track of the JPEG data in the camera image.
+  virtual CameraImageReader* create_image_reader() = 0;
+  // Connection, camera or web server requests one new JPEG image.
+  virtual void request_image(CameraRequester requester) = 0;
+  // Connection, camera or web server requests a stream of images.
+  virtual void start_stream(CameraRequester requester) = 0;
+  // Connection or web server stops the previously started stream.
+  virtual void stop_stream(CameraRequester requester) = 0;
+  virtual ~Camera() {}
+  /// The singleton instance of the camera implementation.
+  static Camera *global_camera;
+};
+
+}  // namespace camera
+}  // namespace esphome
diff --git a/esphome/components/esp32_camera/__init__.py b/esphome/components/esp32_camera/__init__.py
index 2f1f9b90bb..a4b219767a 100644
--- a/esphome/components/esp32_camera/__init__.py
+++ b/esphome/components/esp32_camera/__init__.py
@@ -291,7 +291,7 @@ async def to_code(config):
     cg.add(var.set_frame_buffer_count(config[CONF_FRAME_BUFFER_COUNT]))
     cg.add(var.set_frame_size(config[CONF_RESOLUTION]))
 
-    cg.add_define("USE_ESP32_CAMERA")
+    cg.add_define("USE_CAMERA")
 
     if CORE.using_esp_idf:
         add_idf_component(
diff --git a/esphome/components/esp32_camera/esp32_camera.cpp b/esphome/components/esp32_camera/esp32_camera.cpp
index e9e9d3cffb..a9b8c9f261 100644
--- a/esphome/components/esp32_camera/esp32_camera.cpp
+++ b/esphome/components/esp32_camera/esp32_camera.cpp
@@ -13,8 +13,6 @@ static const char *const TAG = "esp32_camera";
 
 /* ---------------- public API (derivated) ---------------- */
 void ESP32Camera::setup() {
-  global_esp32_camera = this;
-
   /* initialize time to now */
   this->last_update_ = millis();
 
@@ -36,7 +34,7 @@ void ESP32Camera::setup() {
   xTaskCreatePinnedToCore(&ESP32Camera::framebuffer_task,
                           "framebuffer_task",  // name
                           1024,                // stack size
-                          nullptr,             // task pv params
+                          this,             // task pv params
                           1,                   // priority
                           nullptr,             // handle
                           1                    // core
@@ -165,7 +163,7 @@ void ESP32Camera::loop() {
   const uint32_t now = millis();
   if (this->idle_update_interval_ != 0 && now - this->last_idle_request_ > this->idle_update_interval_) {
     this->last_idle_request_ = now;
-    this->request_image(IDLE);
+    this->request_image(camera::IDLE);
   }
 
   // Check if we should fetch a new image
@@ -191,7 +189,7 @@ void ESP32Camera::loop() {
     xQueueSend(this->framebuffer_return_queue_, &fb, portMAX_DELAY);
     return;
   }
-  this->current_image_ = std::make_shared<CameraImage>(fb, this->single_requesters_ | this->stream_requesters_);
+  this->current_image_ = std::make_shared<ESP32CameraImage>(fb, this->single_requesters_ | this->stream_requesters_);
 
   ESP_LOGD(TAG, "Got Image: len=%u", fb->len);
   this->new_image_callback_.call(this->current_image_);
@@ -214,8 +212,6 @@ ESP32Camera::ESP32Camera() {
   this->config_.fb_count = 1;
   this->config_.grab_mode = CAMERA_GRAB_WHEN_EMPTY;
   this->config_.fb_location = CAMERA_FB_IN_PSRAM;
-
-  global_esp32_camera = this;
 }
 
 /* ---------------- setters ---------------- */
@@ -343,7 +339,7 @@ void ESP32Camera::set_frame_buffer_count(uint8_t fb_count) {
 }
 
 /* ---------------- public API (specific) ---------------- */
-void ESP32Camera::add_image_callback(std::function<void(std::shared_ptr<CameraImage>)> &&callback) {
+void ESP32Camera::add_image_callback(std::function<void(std::shared_ptr<camera::CameraImage>)> &&callback) {
   this->new_image_callback_.add(std::move(callback));
 }
 void ESP32Camera::add_stream_start_callback(std::function<void()> &&callback) {
@@ -352,15 +348,18 @@ void ESP32Camera::add_stream_start_callback(std::function<void()> &&callback) {
 void ESP32Camera::add_stream_stop_callback(std::function<void()> &&callback) {
   this->stream_stop_callback_.add(std::move(callback));
 }
-void ESP32Camera::start_stream(CameraRequester requester) {
+camera::CameraImageReader* ESP32Camera::create_image_reader() {
+  return new ESP32CameraImageReader;
+}
+void ESP32Camera::start_stream(camera::CameraRequester requester) {
   this->stream_start_callback_.call();
   this->stream_requesters_ |= (1U << requester);
 }
-void ESP32Camera::stop_stream(CameraRequester requester) {
+void ESP32Camera::stop_stream(camera::CameraRequester requester) {
   this->stream_stop_callback_.call();
   this->stream_requesters_ &= ~(1U << requester);
 }
-void ESP32Camera::request_image(CameraRequester requester) { this->single_requesters_ |= (1U << requester); }
+void ESP32Camera::request_image(camera::CameraRequester requester) { this->single_requesters_ |= (1U << requester); }
 void ESP32Camera::update_camera_parameters() {
   sensor_t *s = esp_camera_sensor_get();
   /* update image */
@@ -389,39 +388,38 @@ void ESP32Camera::update_camera_parameters() {
 bool ESP32Camera::has_requested_image_() const { return this->single_requesters_ || this->stream_requesters_; }
 bool ESP32Camera::can_return_image_() const { return this->current_image_.use_count() == 1; }
 void ESP32Camera::framebuffer_task(void *pv) {
+  ESP32Camera *that = (ESP32Camera *)pv;
   while (true) {
     camera_fb_t *framebuffer = esp_camera_fb_get();
-    xQueueSend(global_esp32_camera->framebuffer_get_queue_, &framebuffer, portMAX_DELAY);
+    xQueueSend(that->framebuffer_get_queue_, &framebuffer, portMAX_DELAY);
     // return is no-op for config with 1 fb
-    xQueueReceive(global_esp32_camera->framebuffer_return_queue_, &framebuffer, portMAX_DELAY);
+    xQueueReceive(that->framebuffer_return_queue_, &framebuffer, portMAX_DELAY);
     esp_camera_fb_return(framebuffer);
   }
 }
 
-ESP32Camera *global_esp32_camera;  // NOLINT(cppcoreguidelines-avoid-non-const-global-variables)
-
 /* ---------------- CameraImageReader class ---------------- */
-void CameraImageReader::set_image(std::shared_ptr<CameraImage> image) {
-  this->image_ = std::move(image);
+void ESP32CameraImageReader::set_image(std::shared_ptr<camera::CameraImage> image) {
+  this->image_ = std::static_pointer_cast<ESP32CameraImage>(image);
   this->offset_ = 0;
 }
-size_t CameraImageReader::available() const {
+size_t ESP32CameraImageReader::available() const {
   if (!this->image_)
     return 0;
 
   return this->image_->get_data_length() - this->offset_;
 }
-void CameraImageReader::return_image() { this->image_.reset(); }
-void CameraImageReader::consume_data(size_t consumed) { this->offset_ += consumed; }
-uint8_t *CameraImageReader::peek_data_buffer() { return this->image_->get_data_buffer() + this->offset_; }
+void ESP32CameraImageReader::return_image() { this->image_.reset(); }
+void ESP32CameraImageReader::consume_data(size_t consumed) { this->offset_ += consumed; }
+uint8_t *ESP32CameraImageReader::peek_data_buffer() { return this->image_->get_data_buffer() + this->offset_; }
 
 /* ---------------- CameraImage class ---------------- */
-CameraImage::CameraImage(camera_fb_t *buffer, uint8_t requesters) : buffer_(buffer), requesters_(requesters) {}
+ESP32CameraImage::ESP32CameraImage(camera_fb_t *buffer, uint8_t requesters) : buffer_(buffer), requesters_(requesters) {}
 
-camera_fb_t *CameraImage::get_raw_buffer() { return this->buffer_; }
-uint8_t *CameraImage::get_data_buffer() { return this->buffer_->buf; }
-size_t CameraImage::get_data_length() { return this->buffer_->len; }
-bool CameraImage::was_requested_by(CameraRequester requester) const {
+camera_fb_t *ESP32CameraImage::get_raw_buffer() { return this->buffer_; }
+uint8_t *ESP32CameraImage::get_data_buffer() { return this->buffer_->buf; }
+size_t ESP32CameraImage::get_data_length() { return this->buffer_->len; }
+bool ESP32CameraImage::was_requested_by(camera::CameraRequester requester) const {
   return (this->requesters_ & (1 << requester)) != 0;
 }
 
diff --git a/esphome/components/esp32_camera/esp32_camera.h b/esphome/components/esp32_camera/esp32_camera.h
index 71f47d3c06..d8f0486ea5 100644
--- a/esphome/components/esp32_camera/esp32_camera.h
+++ b/esphome/components/esp32_camera/esp32_camera.h
@@ -4,7 +4,7 @@
 
 #include "esphome/core/automation.h"
 #include "esphome/core/component.h"
-#include "esphome/core/entity_base.h"
+#include "esphome/components/camera/camera.h"
 #include "esphome/core/helpers.h"
 #include <esp_camera.h>
 #include <freertos/FreeRTOS.h>
@@ -15,9 +15,6 @@ namespace esp32_camera {
 
 class ESP32Camera;
 
-/* ---------------- enum classes ---------------- */
-enum CameraRequester { IDLE, API_REQUESTER, WEB_REQUESTER };
-
 enum ESP32CameraFrameSize {
   ESP32_CAMERA_SIZE_160X120,    // QQVGA
   ESP32_CAMERA_SIZE_176X144,    // QCIF
@@ -73,13 +70,13 @@ enum ESP32SpecialEffect {
 };
 
 /* ---------------- CameraImage class ---------------- */
-class CameraImage {
+class ESP32CameraImage : public camera::CameraImage {
  public:
-  CameraImage(camera_fb_t *buffer, uint8_t requester);
+  ESP32CameraImage(camera_fb_t *buffer, uint8_t requester);
   camera_fb_t *get_raw_buffer();
-  uint8_t *get_data_buffer();
-  size_t get_data_length();
-  bool was_requested_by(CameraRequester requester) const;
+  uint8_t *get_data_buffer() override;
+  size_t get_data_length() override;
+  bool was_requested_by(camera::CameraRequester requester) const override;
 
  protected:
   camera_fb_t *buffer_;
@@ -92,21 +89,21 @@ struct CameraImageData {
 };
 
 /* ---------------- CameraImageReader class ---------------- */
-class CameraImageReader {
+class ESP32CameraImageReader : public camera::CameraImageReader {
  public:
-  void set_image(std::shared_ptr<CameraImage> image);
-  size_t available() const;
-  uint8_t *peek_data_buffer();
-  void consume_data(size_t consumed);
-  void return_image();
+  void set_image(std::shared_ptr<camera::CameraImage> image) override;
+  size_t available() const override;
+  uint8_t *peek_data_buffer() override;
+  void consume_data(size_t consumed) override;
+  void return_image() override;
 
  protected:
-  std::shared_ptr<CameraImage> image_;
+  std::shared_ptr<ESP32CameraImage> image_;
   size_t offset_{0};
 };
 
 /* ---------------- ESP32Camera class ---------------- */
-class ESP32Camera : public Component, public EntityBase {
+class ESP32Camera : public Component, public camera::Camera {
  public:
   ESP32Camera();
 
@@ -155,15 +152,15 @@ class ESP32Camera : public Component, public EntityBase {
   void dump_config() override;
   float get_setup_priority() const override;
   /* public API (specific) */
-  void start_stream(CameraRequester requester);
-  void stop_stream(CameraRequester requester);
-  void request_image(CameraRequester requester);
+  void start_stream(camera::CameraRequester requester);
+  void stop_stream(camera::CameraRequester requester);
+  void request_image(camera::CameraRequester requester);
   void update_camera_parameters();
 
-  void add_image_callback(std::function<void(std::shared_ptr<CameraImage>)> &&callback);
+  void add_image_callback(std::function<void(std::shared_ptr<camera::CameraImage>)> &&callback);
   void add_stream_start_callback(std::function<void()> &&callback);
   void add_stream_stop_callback(std::function<void()> &&callback);
-
+  camera::CameraImageReader* create_image_reader() override;
  protected:
   /* internal methods */
   bool has_requested_image_() const;
@@ -199,12 +196,12 @@ class ESP32Camera : public Component, public EntityBase {
   uint32_t idle_update_interval_{15000};
 
   esp_err_t init_error_{ESP_OK};
-  std::shared_ptr<CameraImage> current_image_;
+  std::shared_ptr<ESP32CameraImage> current_image_;
   uint8_t single_requesters_{0};
   uint8_t stream_requesters_{0};
   QueueHandle_t framebuffer_get_queue_;
   QueueHandle_t framebuffer_return_queue_;
-  CallbackManager<void(std::shared_ptr<CameraImage>)> new_image_callback_{};
+  CallbackManager<void(std::shared_ptr<camera::CameraImage>)> new_image_callback_{};
   CallbackManager<void()> stream_start_callback_{};
   CallbackManager<void()> stream_stop_callback_{};
 
@@ -212,13 +209,10 @@ class ESP32Camera : public Component, public EntityBase {
   uint32_t last_update_{0};
 };
 
-// NOLINTNEXTLINE(cppcoreguidelines-avoid-non-const-global-variables)
-extern ESP32Camera *global_esp32_camera;
-
 class ESP32CameraImageTrigger : public Trigger<CameraImageData> {
  public:
   explicit ESP32CameraImageTrigger(ESP32Camera *parent) {
-    parent->add_image_callback([this](const std::shared_ptr<esp32_camera::CameraImage> &image) {
+    parent->add_image_callback([this](const std::shared_ptr<camera::CameraImage> &image) {
       CameraImageData camera_image_data{};
       camera_image_data.length = image->get_data_length();
       camera_image_data.data = image->get_data_buffer();
diff --git a/esphome/components/esp32_camera_web_server/__init__.py b/esphome/components/esp32_camera_web_server/__init__.py
index d8afea27b4..37598c038b 100644
--- a/esphome/components/esp32_camera_web_server/__init__.py
+++ b/esphome/components/esp32_camera_web_server/__init__.py
@@ -3,7 +3,7 @@ import esphome.codegen as cg
 from esphome.const import CONF_ID, CONF_PORT, CONF_MODE
 
 CODEOWNERS = ["@ayufan"]
-DEPENDENCIES = ["esp32_camera"]
+DEPENDENCIES = ["camera"]
 MULTI_CONF = True
 
 esp32_camera_web_server_ns = cg.esphome_ns.namespace("esp32_camera_web_server")
diff --git a/esphome/components/esp32_camera_web_server/camera_web_server.cpp b/esphome/components/esp32_camera_web_server/camera_web_server.cpp
index 7ca0c56d23..547a5c5361 100644
--- a/esphome/components/esp32_camera_web_server/camera_web_server.cpp
+++ b/esphome/components/esp32_camera_web_server/camera_web_server.cpp
@@ -40,7 +40,7 @@ CameraWebServer::CameraWebServer() {}
 CameraWebServer::~CameraWebServer() {}
 
 void CameraWebServer::setup() {
-  if (!esp32_camera::global_esp32_camera || esp32_camera::global_esp32_camera->is_failed()) {
+  if (!camera::Camera::global_camera) {
     this->mark_failed();
     return;
   }
@@ -67,8 +67,8 @@ void CameraWebServer::setup() {
 
   httpd_register_uri_handler(this->httpd_, &uri);
 
-  esp32_camera::global_esp32_camera->add_image_callback([this](std::shared_ptr<esp32_camera::CameraImage> image) {
-    if (this->running_ && image->was_requested_by(esp32_camera::WEB_REQUESTER)) {
+  camera::Camera::global_camera->add_image_callback([this](std::shared_ptr<camera::CameraImage> image) {
+    if (this->running_ && image->was_requested_by(camera::WEB_REQUESTER)) {
       this->image_ = std::move(image);
       xSemaphoreGive(this->semaphore_);
     }
@@ -106,8 +106,8 @@ void CameraWebServer::loop() {
   }
 }
 
-std::shared_ptr<esphome::esp32_camera::CameraImage> CameraWebServer::wait_for_image_() {
-  std::shared_ptr<esphome::esp32_camera::CameraImage> image;
+std::shared_ptr<esphome::camera::CameraImage> CameraWebServer::wait_for_image_() {
+  std::shared_ptr<esphome::camera::CameraImage> image;
   image.swap(this->image_);
 
   if (!image) {
@@ -170,7 +170,7 @@ esp_err_t CameraWebServer::streaming_handler_(struct httpd_req *req) {
   uint32_t last_frame = millis();
   uint32_t frames = 0;
 
-  esp32_camera::global_esp32_camera->start_stream(esphome::esp32_camera::WEB_REQUESTER);
+  camera::Camera::global_camera->start_stream(esphome::camera::WEB_REQUESTER);
 
   while (res == ESP_OK && this->running_) {
     auto image = this->wait_for_image_();
@@ -203,7 +203,7 @@ esp_err_t CameraWebServer::streaming_handler_(struct httpd_req *req) {
     res = httpd_send_all(req, STREAM_ERROR, strlen(STREAM_ERROR));
   }
 
-  esp32_camera::global_esp32_camera->stop_stream(esphome::esp32_camera::WEB_REQUESTER);
+  camera::Camera::global_camera->stop_stream(esphome::camera::WEB_REQUESTER);
 
   ESP_LOGI(TAG, "STREAM: closed. Frames: %" PRIu32, frames);
 
@@ -213,7 +213,7 @@ esp_err_t CameraWebServer::streaming_handler_(struct httpd_req *req) {
 esp_err_t CameraWebServer::snapshot_handler_(struct httpd_req *req) {
   esp_err_t res = ESP_OK;
 
-  esp32_camera::global_esp32_camera->request_image(esphome::esp32_camera::WEB_REQUESTER);
+  camera::Camera::global_camera->request_image(esphome::camera::WEB_REQUESTER);
 
   auto image = this->wait_for_image_();
 
diff --git a/esphome/components/esp32_camera_web_server/camera_web_server.h b/esphome/components/esp32_camera_web_server/camera_web_server.h
index f65625554c..99c816de27 100644
--- a/esphome/components/esp32_camera_web_server/camera_web_server.h
+++ b/esphome/components/esp32_camera_web_server/camera_web_server.h
@@ -6,7 +6,7 @@
 #include <freertos/FreeRTOS.h>
 #include <freertos/semphr.h>
 
-#include "esphome/components/esp32_camera/esp32_camera.h"
+#include "esphome/components/camera/camera.h"
 #include "esphome/core/component.h"
 #include "esphome/core/helpers.h"
 #include "esphome/core/preferences.h"
@@ -32,7 +32,7 @@ class CameraWebServer : public Component {
   void loop() override;
 
  protected:
-  std::shared_ptr<esphome::esp32_camera::CameraImage> wait_for_image_();
+  std::shared_ptr<camera::CameraImage> wait_for_image_();
   esp_err_t handler_(struct httpd_req *req);
   esp_err_t streaming_handler_(struct httpd_req *req);
   esp_err_t snapshot_handler_(struct httpd_req *req);
@@ -40,7 +40,7 @@ class CameraWebServer : public Component {
   uint16_t port_{0};
   void *httpd_{nullptr};
   SemaphoreHandle_t semaphore_;
-  std::shared_ptr<esphome::esp32_camera::CameraImage> image_;
+  std::shared_ptr<camera::CameraImage> image_;
   bool running_{false};
   Mode mode_{STREAM};
 };
diff --git a/esphome/core/component_iterator.cpp b/esphome/core/component_iterator.cpp
index da593340c1..c68daa04e5 100644
--- a/esphome/core/component_iterator.cpp
+++ b/esphome/core/component_iterator.cpp
@@ -158,16 +158,16 @@ void ComponentIterator::advance() {
       }
       break;
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
     case IteratorState::CAMERA:
-      if (esp32_camera::global_esp32_camera == nullptr) {
+      if (camera::Camera::global_camera == nullptr) {
         advance_platform = true;
       } else {
-        if (esp32_camera::global_esp32_camera->is_internal() && !this->include_internal_) {
+        if (camera::Camera::global_camera->is_internal() && !this->include_internal_) {
           advance_platform = success = true;
           break;
         } else {
-          advance_platform = success = this->on_camera(esp32_camera::global_esp32_camera);
+          advance_platform = success = this->on_camera(camera::Camera::global_camera);
         }
       }
       break;
@@ -386,8 +386,8 @@ bool ComponentIterator::on_begin() { return true; }
 #ifdef USE_API
 bool ComponentIterator::on_service(api::UserServiceDescriptor *service) { return true; }
 #endif
-#ifdef USE_ESP32_CAMERA
-bool ComponentIterator::on_camera(esp32_camera::ESP32Camera *camera) { return true; }
+#ifdef USE_CAMERA
+bool ComponentIterator::on_camera(camera::Camera *camera) { return true; }
 #endif
 #ifdef USE_MEDIA_PLAYER
 bool ComponentIterator::on_media_player(media_player::MediaPlayer *media_player) { return true; }
diff --git a/esphome/core/component_iterator.h b/esphome/core/component_iterator.h
index 9e187f6c57..47ba77e249 100644
--- a/esphome/core/component_iterator.h
+++ b/esphome/core/component_iterator.h
@@ -4,8 +4,8 @@
 #include "esphome/core/controller.h"
 #include "esphome/core/helpers.h"
 
-#ifdef USE_ESP32_CAMERA
-#include "esphome/components/esp32_camera/esp32_camera.h"
+#ifdef USE_CAMERA
+#include "esphome/components/camera/camera.h"
 #endif
 
 namespace esphome {
@@ -48,8 +48,8 @@ class ComponentIterator {
 #ifdef USE_API
   virtual bool on_service(api::UserServiceDescriptor *service);
 #endif
-#ifdef USE_ESP32_CAMERA
-  virtual bool on_camera(esp32_camera::ESP32Camera *camera);
+#ifdef USE_CAMERA
+  virtual bool on_camera(camera::Camera *camera);
 #endif
 #ifdef USE_CLIMATE
   virtual bool on_climate(climate::Climate *climate) = 0;
@@ -123,7 +123,7 @@ class ComponentIterator {
 #ifdef USE_API
     SERVICE,
 #endif
-#ifdef USE_ESP32_CAMERA
+#ifdef USE_CAMERA
     CAMERA,
 #endif
 #ifdef USE_CLIMATE
diff --git a/esphome/core/defines.h b/esphome/core/defines.h
index b5511b57eb..1655a54c00 100644
--- a/esphome/core/defines.h
+++ b/esphome/core/defines.h
@@ -102,7 +102,7 @@
 #define USE_ESP32_BLE
 #define USE_ESP32_BLE_CLIENT
 #define USE_ESP32_BLE_SERVER
-#define USE_ESP32_CAMERA
+#define USE_CAMERA
 #define USE_IMPROV
 #define USE_MICRO_WAKE_WORD_VAD
 #define USE_MICROPHONE